Pacific Spaces: Translations and Transmutations 9781800736269

Delving into Pacific spaces from a variety of disciplinary perspectives and interpretations, this book looks at how the

2,363 85 3MB

English Pages 218 Year 2022

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Contents
Illustrations
Introduction. Pacific Spaces: Dialogues between Architecture and Anthropology
1 Māori ‘Architectural Anthropology’
2 The Junction of the Tala and the Itu
3 The Energetics of Vā and the Samoan Faletele
4 Vā and Its Relationship to the Samoan Built Environment
5 ‘Carving Costs Nothing’ Māori Woodcarvers Train Wage-Labourers How to Show Up to Work on Time
6 Zombie Architecture: Sacrifice in Pre-contact Polynesian and Classical European Buildings
7 Maunawila Heiau: A Sacred Hawaiian Tempo-Spatial Structure Linking Hawai‘i and Moana Nui
8 Aelon Kein Ad: A Case Study of Rimajol Place Identity in the United States
9 Hinemihi 2.0 Whare-for-Export
10 Travelling Houses: Translation, Change and Ambivalence
Conclusion. Vā: What Is In-Between Architecture and Anthropology?
Glossary
Index
Recommend Papers

Pacific Spaces: Translations and Transmutations
 9781800736269

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Pacific Spaces

Pacific Perspectives

Studies of the European Society for Oceanists

General Editors: Edvard Hviding, University of Bergen Toon van Meijl, Radboud University

Oceania is of enduring contemporary significance in global trajectories of history, politics, economy and ecology, and has remained influential for diverse approaches to studying and understanding human life worlds. The books published in this series explore Oceanic values and imaginations, documenting the unique position of the Pacific region – its cultural and linguistic diversity, its ecological and geographical distinctness and always fascinating experiments with social formations. This series thus conveys the political, economic and moral alternatives that Oceania offers the contemporary world. Volume 10 PACIFIC SPACES Translations and Transmutations Edited by A.-Chr. Engels-Schwarzpaul, Lana Lopesi and Albert L. Refiti

Volume 5 IN THE ABSENCE OF THE GIFT New Forms of Value and Personhood in a Papua New Guinea Community Anders Emil Rasmussen

Volume 9 ENGAGING ENVIRONMENTS IN TONGA Cultivating Beauty and Nurturing Relations in a Changing World Arne Aleksej Perminow

Volume 4 LIVING KINSHIP IN THE PACIFIC Edited by Christina Toren and Simonne Pauwels

Volume 8 REVEALING THE INVISIBLE MINE Social Complexities of an Undeveloped Mining Project Emilia E. Skrzypek Volume 7 IF EVERYONE RETURNED, THE ISLAND WOULD SINK Urbanisation and Migration in Vanuatu Kirstie Petrou Volume 6 PACIFIC REALITIES Changing Perspectives on Resilience and Resistance Edited by Laurent Dousset and Mélissa Nayral

Volume 3 BELONGING IN OCEANIA Movement, Place-Making and Multiple Identifications Edited by Elfriede Hermann, Wolfgang Kempf and Toon van Meijl Volume 2 PACIFIC FUTURES Projects, Politics and Interests Edited by Will Rollason Volume 1 THE ETHNOGRAPHIC EXPERIMENT A.M. Hocart and W.H.R. Rivers in Island Melanesia, 1908 Edited by Edvard Hviding and Cato Berg

Pacific Spaces Translations and Transmutations ♦l♦

Edited by A.-Chr. Engels-Schwarzpaul, Lana Lopesi and Albert L. Refiti

berghahn NEW YORK • OXFORD www.berghahnbooks.com

First published in 2023 by Berghahn Books www.berghahnbooks.com © 2023 A.-Chr. Engels-Schwarzpaul, Lana Lopesi and Albert L. Refiti All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher. Library of Congress Cataloging-in-Publication Data Names: Engels-Schwarzpaul, Anna-Chr., editor. Title: Pacific spaces : translations and transmutations / edited by A.-Chr. Engels-Schwarzpaul, Lana Lopesi and Albert L. Refiti. Description: [New York] : Berghahn Books, 2023. | Series: Pacific perspectives: studies of the European society for oceanists ; volume 10 | Includes bibliographical references and index. Identifiers: LCCN 2022019409 (print) | LCCN 2022019410 (ebook) | ISBN 9781800736252 (Hardback) | ISBN 9781800736269 (eBook) Subjects: LCSH: Public spaces--Islands of the Pacific. | Ethnology--Islands of the Pacific. | Islands of the Pacific--Social life and customs. Classification: LCC NA9053.S6 P33 2023 (print) | LCC NA9053.S6 (ebook) | DDC 720.1/03--dc23/eng/20220718 LC record available at https://lccn.loc.gov/2022019409 LC ebook record available at https://lccn.loc.gov/2022019410 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 978-1-80073-625-2 hardback ISBN 978-1-80073-626-9 ebook https://doi.org/10.3167/9781800736252

Contents ♦l♦

List of Illustrations Introduction Pacific Spaces: Dialogues between Architecture and Anthropology Lana Lopesi, A.-Chr. Engels-Schwarzpaul and Albert L. Refiti

vii 1

1 Māori ‘Architectural Anthropology’ Deidre Brown

11

2 The Junction of the Tala and the Itu Athol Greentree

24

3 The Energetics of Vā and the Samoan Faletele I‘uogafa Tuagalu

37

4 Vā and Its Relationship to the Samoan Built Environment Anne E. Allen

54

5 ‘Carving Costs Nothing’: Māori Woodcarvers Train Wage-Labourers How to Show Up to Work on Time Jacob Culbertson

75

6 Zombie Architecture: Sacrifice in Pre-contact Polynesian and Classical European Buildings Ross Jenner and Albert L. Refiti

96

7 Maunawila Heiau: A Sacred Hawaiian Tempo-Spatial Structure Linking Hawai‘i and Moana Nui Tēvita O. Ka‘ili

113

8 Aelon Kein Ad: A Case Study of Rimajol Place Identity in the United States James Miller

129

9 Hinemihi 2.0: Whare-for-Export Anthony Hoete

148

vi  Contents

10 Travelling Houses: Translation, Change and Ambivalence A.-Chr. Engels-Schwarzpaul

159

Conclusion Vā: What Is In-Between Architecture and Anthropology? Albert L. Refiti and A.-Chr. Engels-Schwarzpaul

184

Glossary

195

Index

203

Illustrations ♦l♦

Figures Figure 1.1. Installation of waka taua and pātaka in the new Dominion Museum (now known as the Museum of New Zealand Te Papa Tongarewa), Wellington. The alcove for the Te Hau-ki-Tūranga meeting house can be seen in the right-hand wall. Weekly News, 3 June 1936, front cover. © Sir George Grey Special Collection, Auckland Libraries, AWNS-19360603-39-2. Figure 1.2. Plan illustrating the binary opposites associated with the social usages and meaning of a Māori meeting house. In Anne Salmond, Hui: A Study of Maori Ceremonial Gatherings (Auckland: Reed Methuen, 1975). © Anne Salmond. Figure 2.1. Fale interior and construction of a tala (rounded end). Samoan Material Culture, New Zealand Electronic Text Collection, Victoria University of Wellington. © Te Rangi Hiroa. CC A-SA 3.0. Figure 2.2. Typical fale sections. A, B and C depict three main points of the junction; A = moamoa; B = ulu‘aso and fatuga; and C = amopou and faulalo. © Athol Greentree. Figure 2.3. In Buck (1930), Faulalo to Amopou junction – internal (b) and external (c) views. © Samoan Material Culture, New Zealand Electronic Text Collection, Victoria University of Wellington. CC A-SA 3.0. Figure 2.4. The construction detailing of eaves in metal clad roofs necessitates a constant level ring beam. © Athol Greentree. Figure 2.5. Ulu‘aso lashed to the underside of the fatuga batten. © Athol Greentree.

12

15

25 27

29 30 33

viii  Illustrations

Figure 3.1. Flows of mana energy from Lagi through the faletele and the assembled chefs to the village. © Albert L. Refiti. 44 Figure 4.1. The linear hierarchies in fono seating as prestige vectors and nodes. © Anne E. Allen. 59 Figure 4.2. Topographical ideogram of prestige in the vā established by seating positions. © Anne E. Allen. 64 Figure 6.1. (a) The Parthenon, the Theatre of Herodes Atticus in the front, and the Propylaia (the entrance to the Acropolis) in the top left. Photo by Larry from Charlottetown, PEI, Canada, 2010. CC 2.0. (b)  Youths leading a heifer to the sacrifice. Block XLIV (fig. 132–136) from the South frieze of the Parthenon, ca. 447–433 BC. CC 2.5. 100 Figure 6.2. (a)  The Origin of the Corinthian Order, engraving, 1684, from Claude Perrault’s French translation of Vitruvius. Transferred from en.wikipedia to Commons by Markellos using CommonsHelper, author unknown. (b)  Zeus altar in Olympia, author Davide Mauro, 2018. CC 4.0. 102 Figure 7.1. Ahu‘ena Heiau, Kona, Hawai‘i. Dedicated to Lono. © Tēvita O. Ka‘ili. 116 Figure 7.2. Maunawila Heiau is celestially aligned. Photo by University of Hawai‘i Mānoa Geography Department. Lines and labels of solstices and equinoxes added by author. © Tēvita O. Ka‘ili. 119 Figure 8.1. Map of Marshallese communities across the United States. Tacks represent known Marshallese communities. This living map was created and maintained by Carmen Chong Gum, the former Marshallese diplomat at the Springdale, Arkansas consulate. © James Miller. 131 Figure 8.2. Illustration of Deacon’s home layout demonstrating the diverse uses of spaces within the dwelling. Dwelling size is 120 metres squared. © James Miller and Purvanig Patel. 138 Figure 9.1. WHAT_architecture depiction of Hinemihi envisaged without ornament. © Anthony Hoete and WHAT_architecture. 151 Figure 9.2. Marae-as-whare: proposal for Hinemihi 2.0 integrating wharepuni (sleeping), wharekai (eating, dining) and wharepaku (toilet). © Anthony Hoete and WHAT_architecture. 155





Illustrations  ix

Table Table 8.1.

List of culturally supportive, Rimajol socio-spatial patterns of habitation. © James Miller.

135

Introduction

Pacific Spaces Dialogues between Architecture and Anthropology ♦l♦

Lana Lopesi, A.-Chr. Engels-Schwarzpaul and Albert L. Refiti

T

he chapters in this volume travelled an undulating path to arrive at this point. Beginning in 2006, with our discovery of the Tropical Islands Resort in Germany, and a Samoan fale (traditional house) produced for their Tropical Village, we developed an interest in the relationships of Pacific spaces and buildings with different communities and locations. We formalized this interest in a research project, Travelling Houses, and ensuing discussions with other artists and scholars in the field, like Salā Lemi Ponifasio, Aiono Fana‘afi Le Tagaloa and Hūfanga-He-Ako-MoeLotu ‘Ōkusitino Māhina. This continued in 2014 with extended talanoa (dialogue) which brought together a mixture of architects, cultural studies theorists, artists and anthropologists from across the Moana (Pacific),1 the United States and Europe. In 2015, Leasiolagi Malama Meleisea offered the project some sage advice, stressing that Pacific architecture needed some attention. While the engagement of Indigenous Moana concepts such as mana (prestige, authority, power, influence), tapu (sacred, prohibited, restricted) and noa (ordinary, unrestricted) for the creation and maintenance of space, particularly in contemporary buildings, were of great interest initially, we found that participants also raised questions concerning ecology, creativity, transpersonal wayfinding, the relationships between openness and

2  Lana Lopesi, A.-Chr. Engels-Schwarzpaul and Albert L. Refiti



thresholds, fluidity and control in social space, and cross-cultural global encounters – all of which sit at the intersection of architecture and anthropology. This became even clearer during the following year, when Paige West noted a convergence between architecture and anthropology in our discussions and considered this convergence one of the most interesting and novel aspects of the collaboration. This was perhaps not coincidental since this orientation aligns generally with the work of our Vā Moana – Pacific Spaces research cluster at Auckland University of Technology (then called Pacific Spaces). The cluster provides an international research platform, engaging both Moana and Western thought, to investigate Pacific Space, or Vā Moana. Our aspiration is to transform current thinking about contemporary and customary Indigenous Moana ways of knowing the world, by examining modes of producing space, objects, rituals and performance. How do Moana concepts affect the production of space and the use of the built environment? Moana people have always produced novel ways of understanding the world, based on and extending tradition – how does this relationship play out in contemporary environments? These collaborations and discussions over the past decade set our own, primarily spatially and performatively inflected lens in relationship with anthropological forms of exploration of the same spaces. Paige West in her comments referred to Povinelli (2001: 320), who argues that if ‘indeterminacy refers to the possibility of describing a phenomenon in two or more equally true ways, then incommensurability refers to a state in which two phenomena (or worlds) cannot be compared by a third without producing serious distortion’. This observation prompted Paige to query how we read ‘spaces, affects, relations (and possibilities of relations), and what kinds of lens different kinds of theoretical abstractions bring to our thinking’ (West 2016). Each paper in the talanoa, she noted, asked the reader to think about a phenomenon in at least two ways, and then ‘use those refractions to tell us something about both the phenomenon and how differently situated humans experience the phenomenon’. The conceptual observations and reflections between anthropology and architecture the contributors made ‘about how space, affect, and relations are and are becoming across the Pacific’, she thought, constituted the strength of the collection. Relations, in her words, are the ‘various forms of exchange, interaction, and co-production between beings, forms, objects, and processes as they are understood through multiple ontologies’. In refining the chapters for this publication, we were guided in our decisions by these considerations. Our editorial stance was refined by one of the reviewers, who suggested that the ‘across’ mentioned by Paige West spans multiple disciplines, not just architecture and anthropology. This is generally not surprising. Many Moana researchers do not necessarily



Introduction  3

identify and align themselves with Western disciplinary areas and methods (Wood 2003: 388), and that quite often leads to innovative strategies and transformation of genres. Some examples here in Aotearoa are Albert Wendt, Selwyn Muru, Teresia Teaiwa, John Pule, Karlo Mila and Filipe Tohi. Disciplinary confinement, which many Western educated creative practitioners and scholars take as read, is not a given for ‘Indigenous writers and artists, who often work in multiple media and who often juxtapose genres and forms, such as a written poem and a drawing, painting, sculpture, carving, textile, basket, photograph, moving image, or live performance’ (Allen 2012: xxii). Moreover, when disciplines ‘separate the spiritual from the political, literature from history, or economics from psychology’, they misconstrue how people in Oceania commonly live (Wood 2006: 36) and place an undue pressure for coherence on many interpretations. Transdisciplinary projects come perhaps with a risk, by embracing difficulty, but they can also be more adventuresome and stimulating (Allen 2012: xv), creating an ‘ever-shifting, ever-vibrant space’ for ‘attachments to land and ancestors and … identities formed in experiences of travel, relocation, and dislocation’ (Wood 2003: 388). This collection comes to Pacific spaces from a variety of disciplinary perspectives and interpretations to think through disciplinary translations between anthropology and architectural practice, mostly in the context of the transportation of Pacific spaces. One reviewer suggested that we change our title to Pacific Spaces: Translations and Transmutations. In adopting this suggestion, we thought more about the implications of the particle ‘trans-’ – as in the transnational groups in which people live outside their homelands. Its etymological roots imply ‘crossings’, which resonates with voyaging, but also ‘overcoming’, perhaps as in adversities or difficulties. In science, ‘trans-’ can designate a compound of two distinct chemical groups. For Chadwick Allen, ‘trans-’ enables the augmenting and expanding of broader fields of enquiry. If comparative projects focus on ‘trans-’ rather than ‘ands’, they may hold in balance ‘the complex, contingent asymmetry and the potential risks of unequal encounters’ across different fields and indicate specific agencies and interactions (Allen 2012: xiv). As a metaphor to think with, ‘trans-’ invites a broader spectrum of comparisons between pairs of distinct concepts that lies beyond the scope of this collection. Suffice it to say that thinking of the world in pairings of hoa or soa (Māori or Samoan/ Tongan: friend, companion, pair) underpins, as Tēvita Ka‘ili observed in his keynote at the Interstices Under Construction Symposium in 2017, Moana philosophy.2 Hoa/soa implies notions of ‘similarity, complementarity, duality, equality, tension, conflict’ (Ka‘ili 2017: 9:51) and seems to resonate with comparisons performed under notions of ‘trans-’. We can only make a start

4  Lana Lopesi, A.-Chr. Engels-Schwarzpaul and Albert L. Refiti



here, and it would be interesting to see, as the other reviewer noted, what positions, perspectives and exchanges are possible from perspectives including more and other Moana communities. Transindigeneity, writes Ka‘ili (see Chapter 7), ‘focuses on the strategic juxtapositioning of Indigenous traditions, knowledge, histories, arts, literature, and orature to see what insights, cultural truths, and possibilities … might emerge’. For this purpose, translations are required, often even between groups using the same or similar language. Translation is a relationship that involves everyone, an opening for dialogue, working across knowledges and positions. It is ‘a putting in touch with’, or nothing (Berman 1992: 4). To bring concepts and terms from different traditions together can be a first step in building relationships (see Engels-Schwarzpaul, Chapter 10). In the context of this collection, translation offers a way to understand how (spatial) distance operates in the travels of Pacific spaces, literally or metaphorically. It also helps to explain the acts performed when moving between disciplines. The contributors to this book – architectural practitioners, architectural and spatial design theorists, anthropologists and historians – show not only how new theoretical perspectives can arise out of the rubbing of aspects specific to one discipline against their equivalents of another. They also demonstrate how, in a juxtaposition of terms and concepts proper to Moana ontologies and epistemologies, on one hand, and their corresponding counterparts in Western knowledge traditions, on the other, a space of emergence is created for something that goes beyond both, enhancing both fields of potentialities. We will say more about the development of this space in the conclusion. Rather than attempting direct translations between the disciplines and cultures, most contributors engage, sometimes obliquely, with the question of how buildings corral and hold communities and their rituals together in the Pacific. Preserving specific cultural and social positions throughout multiple readings of sites or buildings, they unfurl a shared horizon within which to re-explore Pacific concepts related to mana and tapu, whakapapa (genealogy) or mafua‘aga (origin), vā and vā fealoaloa‘i (relational space and social relationships), kaitiakitanga (guardianship) and utu (reciprocity). Such concepts and their attendant actions can mostly be traced back to homeland contexts and then extended to encompass recent diasporic situations into which buildings, people and ritual spaces have moved. To reassess contexts and rituals is especially important in the chapters by Pacific scholars, who attempt to construct Indigenous perspectives by drawing on these concepts and the intersections between them. Collectively, the authors consider a broad spectrum of hoa/soa notions derived from Western scholarship, such as traditional – contemporary,



Introduction  5

locally intensive – globally dispersed, relational – singular. They provide fresh insights into the conception, production, designation, articulation and use of physical and cultural states and constructs in the Moana. From the in-between, they suggest different possibilities of thinking about and looking at space and its experience. Further, while all contributors explore spatial relationships in the built environment across the Moana, their observations and arguments also involve strong cultural, geographical and temporal dimensions. The changes that occur in the spatio-temporal movements between people and place, architecture and use, are partly transmutations initiated by coincidental developments in other spheres, or deliberate translations of concepts across different architectural and social traditions. Writing and thinking from the in-between, or vā (Samoan/Tongan/ Fijian: space between), positions this collection within a unique network of researchers from several discipline areas. Early discussions, mostly at ASAO (Association for Social Anthropology in Oceania) meetings, were concerned with anthropological approaches, including perhaps particularly what Marilyn Strathern (2015) calls a counterweight – comparison for the sake of understanding one’s own routine assumptions, as a starting point of an enlarged and more precise ‘thinking between or among’ (Allen 2012: xi). A significantly greater number of contributors have been instrumental in the project’s development at different stages than those assembled here. We would like to acknowledge those who previously presented or developed papers with us for this kaupapa (initiative).3 Looking forward, we hope this transdisciplinary approach to conceptualizing Pacific spaces will offer a framework for thinking from the in-between in the future. Our conversations across disciplines and cultures concerning vā moana eventually led to a research project, Vā Moana: Space and Relationality in Pacific Thought, which, at the time of publication, will be nearing its completion.

Chapter Overviews This collection brings together twelve scholars writing and thinking on Pacific spaces from the in-between. Each chapter enacts various translations and transmutations in their conceptualizations of Pacific spaces, expanding current perspectives on how we come to see and know space in the Moana. These ten chapters can be grouped loosely into two sections. The first section addresses disciplinary translations directly demonstrating the productive tensions of moving between modes of practice, and kinds of knowledge. The second section works to answer the question of how Moana notions of space translate and transmute as they move and shift.

6  Lana Lopesi, A.-Chr. Engels-Schwarzpaul and Albert L. Refiti



Deirdre Brown acknowledges the long history of anthropology on Māori architecture. This constellation is now shifting, as more architecturally trained Māori scholars take up the work of architectural anthropology, and more and more university-trained Māori utilize kaupapa Māori methodologies and participate in architectural design and criticism, advancing a decolonization of Māori architecture. A part of a decolonizing practice recognizing ‘Indigenous sovereign rights over their cultures and stories’ is the inclusion of Māori communities. This, Brown contends, has led to a decentralization of anthropological methods, resulting in a richer multi-disciplinary approach. Athol Greentree is one scholar who exemplifies an architectural anthropology from the position of an architectural practitioner. In his chapter, Greentree writes on the critical junction between the itu (gable roof span) and the tala (rounded ends) sections of the roof of Samoan fale. A ‘tectonic layer’ ensures a structurally sound connection between two fundamentally different geometric forms, which also maintains a ‘commitment to mobility’. A focus on construction origins, Greentree argues, would lead to contemporary fale building, in Sāmoa and the diaspora, based on Indigenous construction methods, rather than Western ones that try to be ‘ornamentally Indigenous’. In the following chapter, I‘uogafa Tuagalu introduces the Samoan concept of vā as intertwining social and sacred spatial relationalities. Tuagalu discusses the meaning of vā over time in several dictionaries, as well as different types of vā and their energies. He suggests that vā can be ‘conceived as a field, in which objects are located and wherein vā-forces operate’. Involving concepts developed by Bourdieu, Bennardo, as well as Lehman and Herdrich, Tuagalu probes into the forces which bring objects into relation, those functioning in vā relations (e.g. mana, tapu and alofa meaning love), as well as corresponding vā-energy flows in Samoan fale. Anne Allen returns to her field work material of the early 1990s to translate and re-translate between French philosopher Henri Lefebvre’s spatial theorizations and the Samoan concept of vā. Allen argues that architecture and the human body, and their placement within vā, function as statements of social organization and are active agents in the Samoan socio-political sphere, which can be read through the lens of Lefebvre’s Representations of Space. The translating necessary for this process, according to Allen, generates ‘a useful vocabulary’ for non-Samoans to understand physical and relational spaces in Sāmoa. Jake Culbertson comes to architecture from anthropology. In his chapter, Culbertson constructs his ethnography based on the time he spent practising with the carvers he writes about: he worked in the House of Knots, carving daily, for periods between two weeks and six months, during the



Introduction  7

five years leading up to 2011. Through this practice of carving, Culbertson contends, carving as it exists in the House of Knots goes beyond serving as a source of inspiration for architectural approaches; it is, rather, a relational practice of care with its own lifeforce. Albert L. Refiti and Ross Jenner write between classical European and Pre-contact Polynesian architectural traditions to find corresponding histories of sacrifice. Ritual arrangements were the formal means to accommodate acts of sacrifice and bloodletting, which, as Refiti and Jenner argue, gave rise to the respective architectures’ primary structural elements. This chapter, too, moves between Moana and Western thought to create shared ground through a reflective dialogue. In the contact zones between these two enquiries emerges a particular kind of transmutation, or (as the authors call it) Zombie architecture. The remaining four chapters explore how spaces, forms and materials are imbued with symbolic aspects when people move, or when buildings themselves are moved. Focusing on the ways in which Pacific spaces are practised enables one to ‘think about similarities and differences between diasporic Oceanic groups and those who have remained nearer to their ancestral islands’ (Wood 2006: 46). Mobilities of peoples and spaces is a translation in and of itself but, as the chapters outlined below reveal, often also lead to a transmutation. Vince Diaz and Kehaulani Kauanui’s notions of transportation and transplanting offer James Miller a way to think through diasporic mobilities, specifically Marshallese place-making in Northwest Arkansas and Oregon. Miller explores Marshallese cultural transference in the reshaping and redefining of their immigrant homes, but also of their environment in the Republic of the Marshall Islands. Notions of place-making and transplanting land, Miller suggests, move us away from damage-centred interpretations of Pacific diasporic spaces and toward more generative futures. Tēvita Ka‘ili focuses on early diasporic settlements across the Pacific from 1200 ad, referring to shared names across the Moana as evidence of early Pacific mobilities. Part of this mobility was the introduction of the rectangular-shaped heiau (place of worship) to Hawai‘i. Ka‘ili recounts the deep history and ancient geography of one specific heiau, Maunawila Heiau, which has helped to reconnect ancestral links between Hawai‘i and the rest of the Moana. Remembering these deep connections, Ka‘ili argues, encourages trans-Indigenous connections, which maintain Indigenous particularities while reaching across Indigenous communities. Hinemihi o te Ao Tāwhito, a whare whakairo (carved meeting house), was purchased in Aotearoa New Zealand by William Hillier Onslow in 1891 and dismantled in the following year to be transported to England by sea. Since 1897, Hinemihi has been in Clandon Park, where she has

8  Lana Lopesi, A.-Chr. Engels-Schwarzpaul and Albert L. Refiti



become a spiritual home for Māori in the London diaspora. In 2017, Anthony Hoete, as chairman of Te Maru o Hinemihi, proposed to the UK National Trust to restore Hinemihi by creating replacement carvings for her original size (before it was shortened in Clandon Park by half) and by extending the whare to serve, again, as a whare manaaki (house to care for guests). Anthony Hoete proposes trans-disciplinary collaborations for the co-design of Hinemihi 2.0, a whare-for-export. Hinemihi is an example of what Tina Engels-Schwarzpaul describes as travelling houses. Locating Hinemihi in a wider network of Māori and Samoan houses in Europe and the USA, Engels-Schwarzpaul focuses on aspects of translations and transmutations that are activated when the houses themselves move (or are moved) away from the people. At several levels, Engels-Schwarzpaul addresses epistemological questions regarding translation and transmutations central to this collection. Translation, Engels-Schwarzpaul contends, with its literal sense of moving from one place to another, ‘appropriately mobilizes meaning and expands interpretive frameworks, facilitates multi-perspectival discussions, works around the blind spots in each culture and helps to uncover or articulate new affinities between Pacific and Western types of knowledge’. The intention for this volume was to instigate dialogues in and across different locations around the Pacific, as well as translations between different disciplines, primarily involving anthropologists and people working in architecture. Such dialogues had already been started in the 1960s, but then disappeared a decade or two later. We hope not only to have taken up the loose threads again but to have opened up a space for new conversations – this time including, notably, a majority of Indigenous scholars as authors and addressing a wider range of concerns. The ensuing discussions reflect the complexity and range of this emergent field of research, in celebration of the productive frictions and tensions between. Lana Lopesi is Assistant Professor in Pacific Islander Studies at the University of Oregon and author of False Divides (2018) and Bloody Woman (2021). Currently she is Editor-in-Chief for the Pacific Art Legacy Project, Editor for Marinade: Aotearoa Journal of Moana Art and Arts Editor, Metro Magazine. She received her PhD in 2021 from Auckland University of Technology on Moana Cosmopolitan Imaginaries. Lana is a researcher for the Vā Moana – Pacific Spaces research cluster at Auckland University of Technology. A.-Chr. Engels-Schwarzpaul is Professor in Spatial Design and Postgraduate Studies at Auckland University of Technology – Te Wānanga Aronui o Tāmaki Makau Rau, in Auckland, Aotearoa/New Zealand. She was



Introduction  9

Executive Editor of Interstices: Journal of Architecture and Related Arts from 2005 to 2012 and edited around eight special issues for Interstices, ACCESS and Knowledge Cultures. She also edited Of Other Thoughts: Non-traditional Ways to the Doctorate: A Guidebook for Candidates and Supervisors (2013, for Sense Publishers, with M.A. Peters) and edited and translated Atmospheric Architectures: The Aesthetics of Felt Spaces  (2017, for Bloomsbury, author Gernot Böhme). Albert L. Refiti is a research leader in the field of Pacific spatial and architectural environment with an extensive research and publication history in the area, supported by his teaching and lecturing in the last fifteen years. His current research is on Pacific concepts of space, how they are formulated and enacted, the aim of which is to find out how this understanding might play a role in rethinking the ways that Pacific people can create new modes of working and creating new notions of place and citizenship in the diaspora towards a Pacific cosmopolitic.

Notes   1. ‘Moana’ refers to what Europeans later called the ‘Pacific Ocean’. The term has become increasingly accepted as one denoting Moana (Pacific) communities, that is (mostly), Polynesian peoples in their homelands, as well as transnational urban communities outside of those homelands. However, the term is still debated: not all people in the Moana feel included. We use ‘Moana’ here interchangeably with ‘Pacific’, as a currently mostly accepted, but problematic and quite possibly interim term.   2. Many Moana cosmologies revolve around cosmic pairs, as in Aotearoa, Papatūānuku and Ranginui.   3. Fa‘afetai tele lava to: Mike Austin, Michael Goldsmith, Jeremy Treadwell, Sean Mallon, Karlo Mila, Moana Nepia, Bruce Moa, Semisi Potauaine, Tate LeFevre, Marianne George, Spencer Leineweber, Sa‘iliemanu Lilomaiava-Doktor, Karamia Müller, Vaoiva Natapu-Ponton, Ramon Tiatia, Jan Rensel, Alan Howard.

References Allen, Chadwick. 2012. Trans-indigenous: Methodologies for Global Literary Studies. Minneapolis: University of Minnesota Press. Berman, Antoine. 1992. The Experience of the Foreign: Culture and Translation in Romantic Germany (trans. S. Heyvaert). Albany, NY: State University of New York Press.

10  Lana Lopesi, A.-Chr. Engels-Schwarzpaul and Albert L. Refiti



Ka‘ili, Tēvita. O. 2017. ‘Pacific Patterns/Moana Philosophy’ [Keynote, 4 June], Interstices Under Construction Symposium. Auckland: Auckland University of Technology. Retrieved 25 April 2022 from https://vimeo.com/221972588. Povinelli, Elizabeth. 2001. ‘Radical Worlds: The Anthropology of Incommensurability and Inconceivability’, Annual Review of Anthropology 30: 319–34. Strathern, Marilyn. 2015. Learning to See in Melanesia: Four Lectures Given in the Department of Social Anthropology, Cambridge University, 1993–2008. Chicago, IL: Hau Books. West, Paige. 2016. Written report on our 2016 ASAO working session in February 2016. Wood, Houston. 2003. ‘Cultural Studies for Oceania’, The Contemporary Pacific 15(2): 340–74. . 2006. ‘Three Competing Research Perspectives for Oceania’, The Contemporary Pacific 18(1): 33–55.

1

Māori ‘Architectural Anthropology’ ♦l♦

Deidre Brown

Introduction

T

he anthropology of Māori architecture has a long academic history. The influence of anthropologists on this subject, however, is lessening as architecturally trained scholars claim this territory, employing methodologies informed by anthropology and other disciplines. This chapter acknowledges the contribution of anthropologists, from Āpirana Ngata’s needs-based analyses, to Anne Salmond’s structuralist interpretations, to Roger Neich’s exacting investigations. It examines the decline of anthropological authority over the ‘field’ of Māori architecture in recent years, as other scholars have actively wrested it into Indigenous knowledges and architectural paradigms. The Māori cultural renaissance from the 1970s, the rise of Māori art and architectural history in the 1980s, and the 2006 amendment to the Antiquities Act (now known as the Protected Objects Act) are influential events leading to this change. Their importance is demonstrated by comparing who is now writing about Māori architecture and Pacific architecture. As this chapter argues, the decolonization of Māori architecture has led to a decentralization of anthropological methods, resulting in a richer multi-disciplinary approach.

Early Ethnology and Anthropology Ethnology had an especially early start in Aotearoa New Zealand, beginning with the speculations on Māori origins by James Cook and Joseph

12  Deidre Brown



Banks in 1769. The Pacific quickly became the great intellectual laboratory of the mid- and late eighteenth century. Exploring expeditions from Britain, France and North America were ‘staffed by many of the leading savants of the day’, wrote the historian M.P.K. Sorrenson (1992: 13), but it was the colonial settlers who founded an enduring local anthropology in Aotearoa New Zealand. Learned Societies and Mechanics Institutes were established from the 1850s by middle-class intellectuals keen to integrate the knowledge of Māori into their own traditions of scholarship. By the 1860s, some of these societies and their scholars had established museums. For example, Julius von Haast, founder of the Philosophical Institute of Canterbury in 1862, became the director of the institute’s Canterbury Museum, six years later. The Museum of New Zealand Te Papa Tongarewa (Wellington) and Auckland War Memorial Museum Tāmaki Paenga Hira had similar origins in the 1860s (Sorrenson 1992: 17). No major metropolitan museum was considered complete without a whare whakairo (decorated Māori meeting house), pātaka (carved Māori raised storehouse) and waka taua (carved Māori war canoe) (see Figure 1.1). These items then became the basis for further intellectual enquiry, despite sometimes being

Figure 1.1 Installation of waka taua and pātaka in the new Dominion Museum (now known as the Museum of New Zealand Te Papa Tongarewa), Wellington. The alcove for the Te Hau-ki-Tūranga meeting house can be seen in the right-hand wall. Weekly News, 3 June 1936, front cover. © Sir George Grey Special Collection, Auckland Libraries, AWNS-19360603-39-2.



Māori ‘Architectural Anthropology’  13

acquired through unscrupulous means. After reports by the Waitangi Tribunal into their acquisition, the whare whakairo accessioned to Otago Museum (Dunedin), Mataatua, was repatriated, reopening in Whakatane in 2011, and Te Hau-ki-Tūranga at Te Papa was recommended for return to the Rongowhakaata tribe (Barrow 1965; Brown 1996; Waitangi Tribunal 1999; 2004). Anthropology did not migrate so quickly to Aotearoa New Zealand. By the late nineteenth century, as Sorrenson (1992: 7) notes, ‘anthropology was still in its infancy and its few professionals were still ensconced in their armchairs in Europe and largely reliant on the communications of amateurs [amateur ethnologists] for the grist of their academic mills’. Even into the next century, the roles of the anthropologist and ethnologist were not easily untangled, since in Aotearoa New Zealand the relatively small research community had less opportunity to specialize. Consequentially, a homegrown ‘architectural anthropology’ has developed, unchallenged. In the early to mid-twentieth century, a few Aotearoa New Zealand scholars attempted to survey the prevalence of certain Māori building ‘types’ according to regional categories, most notably William (W.J.) Phillipps (1944; 1952; 1955; 1956) and James Creswell (1977), but their projects were somewhat defeated by the large number of Māori buildings and the different pūrākau (intergenerational stories, codified knowledges) that could be associated with the same building. The latter issue undermined the search for a singular truth, the benchmark of Western scholarship at the time. Within mātauranga Māori (the Māori intellectual tradition), multiple narratives are an outcome of handing down stories between generations and are equally valued (though fiercely argued by some) within communities. The two different knowledge systems, one based on establishing singular ‘truthful’ narratives and the other concerned with the dynamics of pluralism, have not been reconciled. Neither should they need to be as they come from two different world views (Hikuroa 2017: 9).

Professional Anthropology and Māori Architectural Scholarship Māori architectural knowledge is not concerned with material preservation or conservation (McKay and Walmsley 2003: 85–95; McKay 2004: 295–300; Sully 2007), so much as with continuation of use, an approach understood by Āpirana Ngata, who might be considered a founding figure in architectural anthropology. As a member of Ngāti Porou (one of the few tribes with an unbroken heritage of architectural whakairo rākau (wood carving)), an anthropologist and an influential Member of Parliament (Eastern Māori electorate, 1905–1943), Ngata recognized the social value

14  Deidre Brown



of Māori architecture. He appreciated that the tectonic understanding of building could not be divorced from social custom or self-determination in a colonial age. The School of Māori Arts and Crafts that he founded in 1927 through an Act of Parliament appeared to be an attempt to preserve Māori arts, particularly wood carving. The legislation reiterated the discourse of (Anglo-New Zealand) anthropology: that Māori arts were passing away because of Western influence (Brown 1999: 241–76). This, of course, ensured government support for the School. However, in practice the School was a targeted, needs-based response to the diminishing control that Māori had over their lands due to the individualization of communal land title. The School built large and elaborate whare whakairo and whare kai (dining halls) in the central and eastern North Island. They acted as headquarters and community centres for tribal incorporations, created under the Ngata development legislation, that farmed land on behalf of Māori shareholders. The architecture that had once united hapū (subtribes) was employed in this instance to unite an emerging Māori nation that Ngata hoped would become economically self-sustaining (Brown 1999). Ngata wrote prolifically1 about his ‘scientific’2 revival of tribal carving styles, many of which had declined since European arrival, and their promotion within the School’s whare whakairo projects. Museum-held whakairo rākau and whare whakairo, particularly the Te Hau-ki-Tūranga house in the Dominion Museum (now known as The Museum of New Zealand Te Papa Tongarewa), were ‘studied’ through observation and replication by the School’s tohunga whakairo (carvers), tukutuku (lattice wall-panel) experts and kōwhaiwhai (scroll) painters in order to recover and adapt styles (Brown 1999: 261; Brown 1996). Perhaps mindful that the School might be regarded as a revisionist intervention into, rather than part of the continuum of, Māori architecture, Ngata argued that it perpetuated a practice of copying that had always existed since there had always been ‘a strong tendency to stick to certain features and conventions’ in building and its associated arts (Ngata 1958b: 34). Ngata’s architectural anthropology is too early and too inflected by his training as a classical scholar to be described as the outcome of a kaupapa Māori research and practice methodology, but it certainly was by, with and for Māori, which are often the benchmarks for contemporary Indigenous research.3 It was not until the 1970s that non-Māori scholars began to investigate Māori buildings in collaboration with communities as a way to understand identity and cultural change. In 1975, the Aotearoa New Zealand anthropologist Anne Salmond published her PhD thesis (undertaken at the University of Pennsylvania) under the title Hui: A Study of Maori Ceremonial Gatherings (1975). The book has been reprinted many times



Māori ‘Architectural Anthropology’  15

since. It is an account of how the architecture of the marae (open-air forum) and its buildings (including the whare whakairo and whare kai) is activated by custom, ritual, performance and language during ceremonial occasions. The book’s structuralist pedigree is evident in the employment of binary opposites to describe the meaning of the marae and its buildings, revealing the whare whakairo to be an architectural embodiment of an ancestor, as well as the representation of whakapapa (genealogy) (see Figure 1.2). As the architectural historian Mike Austin found after completing his 1976 PhD (Architecture, University of Auckland) thesis, ‘Polynesian Architecture in New Zealand’, speculative or theoretical comments about spatial arrangements – in his case the relationship between landscape ‘openness and closure’ and the marae formalism – can be appropriated into the source culture, to become one of its many defining narratives. Similarly, the understanding of the marae and meeting house as a set of binary spatial relationships has not been challenged from within the culture.

Figure 1.2. Plan illustrating the binary opposites associated with the social usages and meaning of a Māori meeting house. In Anne Salmond, Hui: A Study of Maori Ceremonial Gatherings (Auckland: Reed Methuen, 1975). © Anne Salmond.

16  Deidre Brown



A point of difference from other anthropological scholarship produced in the decade that Hui was published is the Māori pedagogical mentorship that Salmond received from the Te Whānau-a-Apanui and Ngāti Porou elders Amiria and Eruera Stirling. This is what possibly led her to reflect, in the book’s conclusion (1975: 210), that: The hui [ceremonial gathering on the marae] has a life and validity of its own that is quite independent of anthropology…. The task of the anthropologist is to describe this reality as nearly as possible in its own terms, reflecting its structure in the structure of the description, and remaining faithful to its boundaries and definitions.

Salmond’s own anthropological perspective has changed over the last forty years or so, but the book remains an important foundational text in Māori architectural anthropology, history and theory. In the mid-1980s, David Simmons’ Whakairo: Maori Tribal Art (1985) and Hirini Mead’s Te Toi Whakairo: The Art of Maori Carving (1986) revisited the concept of regional schools of carving, distinguished by tribal affiliations, contextualizing museum-held house components, particularly wood carvings, within continuing traditions of house building. Mead was an anthropologist and Professor of Māori Studies based at Victoria University, Simmons, an anthropologist at Auckland Museum. Each had undertaken considerable field work and investigations of house parts in local and international museums. Because their focus was on whakairo rākau, their books focused on the whare whakairo as a primarily carved object that was ‘authentically’ and ‘traditionally’ Māori. Indeed, Mead (1984: 75) was so philosophically troubled by the emergence of a new generation of contemporary Māori artists and designers, working in the Western idiom with Māori themes, that he wrote: ‘Māori artists trained in the art schools of the Pākehā [New Zealand Europeans] are spearheading a movement to change the face of Māori art more radically than ever before. One does not know whether they innovate with love and understanding, or whether they are about to ignite new fires of destruction’. Simmons and Mead published perhaps the last large scholarly works that situated the whare whakairo within a ‘traditionalized’ context, to use a phrase first coined by the anthropologist Jeffrey Sissons (1998: 36–46) in a paper about the School of Māori Arts and Crafts and its precedents. As Sissons observed, traditionalization was socially important as the architecture reinforced a political system that advocated the restoration of tino rangatiratanga (Māori sovereign rights), but it also purposefully ignored the change, dynamism, development and appropriation inherent in the culture.



Māori ‘Architectural Anthropology’  17

Poststructuralism, Cultural Change and the Rise of Indigenous Scholarship Very little work had been undertaken by anthropologists, ethnologists, historians and architectural scholars into Māori architectural change until, from around 1980, a number of new authors, papers and books began to appear examining the influence of Western culture on Māori culture. Poststructuralism questioned the existence and need for singular truths and authenticity. Major publications in this new vein were produced by the historian Judith Binney, the theologian Bronwyn Elsmore, and Anne Salmond – all of which dealt with the evolving nature of architecture as a consequence of cultural change.4 The most significant contribution to a poststructuralist Māori architectural anthropology was written by Roger Neich in 1993. Painted Histories: Early Māori Figurative Painting was a revised version of Neich’s PhD (Anthropology) from Berkeley, which was itself an outcome of his work as an ethnologist at Aotearoa New Zealand’s National Museum (now Te Papa) in the 1970s and 1980s. It was a detailed account of the development of figurative painting in Māori meeting houses predominantly built in the late nineteenth and early twentieth centuries by members of the Ringatū, a faith that has its origins in both Christianity and Māori spirituality. Likewise, the paintings, and the houses they embellished, owed their aesthetics, composition and technologies to both cultures, although their purpose was for Māori self-determination. Ngata disliked the hybridity of the figurative paintings and their association with the New Zealand Wars and new forms of Māori Christianity, and so, with his encouragement, the art was abandoned. There are nods to structuralist art history and anthropology in the book, as well as poststructuralism, which unbundles information from these disciplines as well as history and architectural history. The impact of Painted Histories cannot be overstated. It informed the work of a new generation of contemporary Māori artists, such as Shane Cotton, Peter Robinson and Chris Heaphy, who identified with the break from convention instigated by the Ringatū artists more than a century before (Tipa 2005: 9). Indeed, it was these artists that Mead was criticizing in 1984. It also validated my own research, and that of Bill McKay’s, into ‘Mōrehu’ architecture – the art of Māori religio-political groups. Most importantly, it returned mana (prestige) to community buildings, and the people who cared for them, that did not conform to the archetype promoted either by the School of Māori Arts and Crafts and its graduates or by earlier ethnographers. In a conversation we had in the mid-1990s, Neich (who passed away in 2010) denied that he was an ‘architectural historian’. The title of ‘architectural anthropologist’ might have been received by him more willingly.

18  Deidre Brown



Running parallel to these developments in anthropology were changes inspired by activism in Indigenous scholarship by Indigenous peoples, and the emergence of architectural history as its own discipline. The 1970s had been a time of intense Māori political action. It was influenced by the American Civil Rights Movement, and embodied by events such as Whina Cooper’s 1975 ‘Not One More Acre of Māori land’ hikoi (land march) from Te Hāpua (at the northernmost end of Aotearoa New Zealand) to Parliament in Wellington, the 1978 police clearance of the Māori occupiers of Takaparawhau (Bastion Point) in Auckland, and the protests against the 1981 Springbok Tour to Aotearoa New Zealand. Although Māori were still a minority on campus, more were beginning to enter university and demand the adoption of culturally appropriate research methodologies by academics working ‘on’ Māori communities. Some of the student activists went on to assume university positions to ensure that change took place. The architectural academy, itself, had experienced a sea change with the acceptance of Austin’s PhD, which demanded that Māori buildings be understood as architecture and not material culture or iterative construction (although the idea took longer to gain traction in the architectural design profession). Internationally, these were the early years of architectural history being taught by studio-trained architects in architecture schools, rather than adjunct art historians. Austin, who is also an architect, taught the Oceanic Architecture course in the first professional year of the architecture programme from the 1980s through to the 1990s at the University of Auckland. Many Aotearoa New Zealand architects who were to self-nominate as bicultural or Indigenous designers were either students of this course, or students of Austin’s graduates. The emergence of an ‘architectural anthropology’ in Aotearoa New Zealand reflects both the easing of anthropology’s boundaries and the adoption of many disciplinary methodologies by scholars of architecture and Indigenous knowledges. Indigenous voices are now more prominent in Māori architectural design, history and criticism. They belong to tertiary-educated academics, artists, architectural designers, art historians, art curators, and curators of ethnology. There are a number of non-Māori academics working in these fields whose authority to speak about Māori architecture comes from their collaborative approach when researching with communities. Their kaupapa Māori research processes have included community-established research questions, co-option of community members as researchers, dissemination through hui (Māori meetings), and action-based outcomes, such as conservation report writing and acting as external ‘experts’ in resource management and heritage procedures. These working methodologies recognize Indigenous sovereign rights over their cultures and stories, and are by no means the standard outside of Māori architectural



Māori ‘Architectural Anthropology’  19

studies. This is evident when comparing who is now writing about Māori art and architecture and Pacific art and architecture. The latter is a much larger academic field, reflecting the vastness of Oceania, but it lacks a critical mass of Indigenous voices. Instead, the discipline is dominated by mostly non-Indigenous anthropologists working with museum- and gallery-based collections that, outside of Aotearoa New Zealand, are largely based in North America and Europe, and away from the influence of the New Zealand Protected Objects Act 1975, which prohibits the export of Māori artefacts. Māori architectural scholarship is increasingly treating buildings as a cultural spatial experience, located in living communities and ancestral landscapes; Pacific architecture, by comparison, has mostly been reduced to parts or objects stored or published or aestheticized within ‘white box’ (art gallery) or ‘black box’ (museum) contexts. The emergence of kaupapa Māori methodologies and a body of vocal Māori academics and community members who resist the commodification of their cultural heritage has been received as a ‘hands off ’ message to international scholars who are not prepared to engage directly with the origin communities of buildings or building parts. As a consequence, there have been only three books on Māori architecture that have originated abroad over the past forty years: Whare Karakia: Māori Church Building, Decoration and Ritual in Aotearoa New Zealand, 1834–1863 by Richard Sundt (an Associate Professor of Art History at the University of Oregon); Decolonizing Conservation: Caring for Māori Meeting Houses outside New Zealand, edited by Dean Sully (a Lecturer in Conservation at University College London); and the Hamburg Museum fur Volkerkunde’s collection of essays (published to coincide with the hundredth anniversary of its acquisition of the whare whakairo Rauru), House Rauru: Masterpiece of the Māori.5 All were produced with collaborative research methodologies and/ or Māori contributors. At home, architectural academics have been selective in what they have appropriated from anthropology, perhaps with the objective in mind of better articulating architectural observations rather than presenting Māori architecture as a case study in world anthropology. Hence, theories of kinship, relationships, social structure, place (especially place attachment), signification and identity have been invoked in published studies undertaken by me, Mike Austin, Bill McKay, Michael Linzey, Jeremy Treadwell and Sarah Treadwell. We have been less enthusiastic about adopting the anthropological concept of ‘custom’ as this suggests that building is an iterative process without innovation, an idea at odds with the purpose of architectural design. There is now a new generation of Māori architectural scholars and practitioners, educated at university and polytechnics under the supervision

20  Deidre Brown



of academics who base their scholarship on methods and knowledges derived from anthropology, Indigenous studies, history and design, as well as the mātauranga Māori of their communities. These emerging scholar practitioners include Rameka Alexander-Tu‘inukuafe, Te Kahuwhero Alexander-Tu‘inukuafe, Ellen Andersen, Savannah Brown, Ambrosia Crum, Aaron Gourlay, Elisapeta Heta, Jade Kake, Matekitātahi RawiriMcdonald (who wrote his Master of Architecture [Professional] thesis in te reo, the Māori language), Nanako McIntosh, Harriet Mildon-King, Amber Ruckes, Raukura Turei and Jessica Young.6 Unlike their predecessors, some in this generation are fluent in te reo and tikanga Māori (Māori customs) due to their upbringing by culturally-conscious parents and their pre-tertiary education in bilingual and Māori immersive language environments. Whereas their teachers took the radical step of looking outside their discipline to appropriate concepts and methods from other fields, including anthropology, these new scholars also move with ease in an expanded professional field that allows them to practise as architectural designers, artists, teachers and writers. Many are also involved in researching their iwi (tribe) and hapū pūrākau as it pertains to architecture, rather than examining architecture from a national ‘Maori’ perspective, which was the project of previous generations of scholars. Consequently, they can directly engage in the self-sustaining, tino rangatiratanga of their communities. Anthropology is one of many different tools they use in their research, at the central core of which is a kaupapa Māori approach.

Conclusion There is much that scholars have to learn, and are learning, from each other in terms of the appreciation and documentation of Indigenous architecture within a social/cultural environment. In his 2007 book on Aboriginal art, Becoming Art, Howard Morphy demonstrated how art history is supported by historiography that has consistently privileged Western perspectives of time, change and continuity. Many Indigenous people also associate anthropology with the categorization and objectification of their culture in Western terms. The challenge for architectural anthropology, like art history, is to define what the discipline is in a new global context that respects and protects Indigenous knowledges. History shows us that architectural anthropology in Aotearoa New Zealand has shifted from being an interpretive to an applied tool that, in the hands of Indigenous scholars over the past century, has been central to the maintenance and promotion of a unique Māori identity. What is emerging is a scholarship that draws on concepts of relativity inherent in contemporary social anthropological discourse, while rejecting the notion that architecture is a fixed iterative



Māori ‘Architectural Anthropology’  21

tradition. Anthropology is also slowly changing from being a profession to a discipline whose core approaches can be appropriated, and in time perhaps modified, by academics from other fields, including architecture. Whether all these academics would call themselves ‘architectural anthropologists’ is another matter. Deidre Brown (Ngāpuhi, Ngāti Kahu) teaches design and history in the School of Architecture at the University of Auckland. Her specialist teaching, supervisory and research interests are in the fields of Māori and Pacific architectural and art history, and the broader discipline of Indigenous design. She has written several books, including the multi-authored Art in Oceania: A New History (2012) and Māori Architecture (2009), and curated a number of exhibitions in galleries around the country.

Notes   1. There are several dozen unpublished documents, memorandums and letters in the Alexander Turnbull Library and National Archives collections by Ngata that document the School’s activities. Published documents are Ngata 1940, 1958a, 1958b.   2. For his discussion of this approach, see Ngata 1940: 329.   3. See Smith 1999.   4. Binney, Chaplin and Wallace 1979; Binney 1995; Binney and Chaplin 1986; Elsmore 1985; Elsmore 1989; Salmond 1991; Salmond 1997.   5. Sundt 2010; Sully 2007; Kopke and Schmelz 2012 (published in English and German).   6. More Māori architectural practitioners / scholars are listed on the Ngā Aho Network of Māori Design Professionals website: http://www.ngaaho.maori. nz/memberResults.php. Women dominate the list.

References Austin, Mike. 1976. ‘Polynesian Architecture in New Zealand’, PhD dissertation. Auckland: University of Auckland. Barrow, Terence. 1965. A Guide to the Meeting House, Te Hau-ki-Turanga. Wellington: Dominion Museum. Binney, Judith. 1995. Redemption Songs: A Life of Te Kooti Arikirangi Te Turuki. Auckland: Auckand University Press and Bridget Williams Books. Binney, Judith and Gillian Chaplin. 1986. Nga Morehu: The Survivors. Auckland: Auckland University Press.

22  Deidre Brown



Binney, Judith, Gillian Chaplin and Craig Wallace. 1979. Mihaia: The Prophet Rua Kenana and His Community at Maungapohatu. Auckland: Auckland University Press. Brown, Deidre. 1996. ‘Te Hau-ki-Turanga’, Journal of the Polynesian Society 105(1): 7–26. . 1997. ‘Moorehu Architecture’, PhD dissertation. Auckland: University of Auckland. . 1999. ‘The Architecture of the School of Maori Arts and Crafts’, Journal of the Polynesian Society 108(3): 241–76. Brown, Deidre and Bill McKay. 1992. ‘Buildings of the Morehu: Te ao hou’, Interstices 2: 112–33. Cresswell, John. 1977. Maori Meeting Houses of the North Island. Auckland: P.C.S. Publishers. Elsmore, Bronwyn. 1985. Like Them That Dream: The Maori and the Old Testament. Tauranga: Tauranga Moana Press. . 1989. Mana from Heaven: A Century of Maori Prophets in New Zealand. Tauranga: Moana Press. Hikuroa, Daniel. 2017. ‘Mātauranga Māori – the Ūkaipō of Knowledge in New Zealand’, Journal of the Royal Society of New Zealand 47(1): 5–10. Kopke, Wulf and Bernd Schmelz. 2012. House Rauru: Masterpiece of the Maori. Hamburg: Museum fur Volkerkunde. McKay, Bill. 2004. ‘Resonant Time and Cyclic Architecture’, in Harriet Edquist and Hélène Frichot (eds), Limits: 21th Annual Conference of the Society of Architectural Historians of Australia and New Zealand. Melbourne: SAHANZ, pp. 295–300. McKay, Bill and Deirdre Brown. 1992. ‘The Architecture of the New Land’, Architecture New Zealand (November/December 1992): 86–89. McKay, Bill and Antonia Walmsley. 2003. ‘Maori Time: Notions of Space, Time and Building Form in the South Pacific’, in Maryam Gusheh and Naomi Stead (eds), Progress: 20th Annual Conference of the Society of Architectural Historians of Australia and New Zealand. Melbourne: SAHANZ, pp. 85–95. Mead, Hirini Sidney Moko. 1984. Te Maori. New York: Harry Abrams Inc. and The American Federation of Arts. . 1986. Te Toi Whakairo: The Art of Maori Carving. Auckand: Reed. Morphy, Howard. 2007. Becoming Art: Exploring Cross-cultural Categories. New York and Oxford: Routledge. Neich, Roger. 1993. Painted Histories: Early Maori Figurative Painting. Auckland: Auckland University Press. Ngata, Apirana. 1940. ‘Maori Arts and Crafts’, in Ivan Lorin George Sutherland (ed.), The Maori People Today. Wellington: New Zealand Institute of International Affairs and the New Zealand Council for Educational Research, pp. 307–35. . 1958a. ‘The Origin of Maori Carving – Part 1’, Te Ao Hou 22: 30–37.



Māori ‘Architectural Anthropology’  23

. 1958b. ‘The Origin of Maori Carving – Part 2’, Te Ao Hou 23: 30–34. Phillipps, William John. 1944. Carved Maori Houses of the Eastern Districts of the North Island. Wellington: Dominion Museum. . 1952. Maori Houses and Food Stores. Wellington: Dominion Museum. . 1955. Carved Maori Houses of the Western and Northern Areas of New Zealand. Wellington: Government Printer. . 1956. The Great Carved House Mataatua of Whakatane. Wellington: Polynesian Society. . n.d. The Te Kuiti House. Wellington: Tombs. Salmond, Anne. 1975. Hui: A Study of Maori Ceremonial Gatherings. Auckland: Reed Methuen. . 1991. Two Worlds: First Meetings Between Maori and Europeans, 1642– 1772. Auckland: Viking. . 1997. Between Worlds: Early Exchanges Between Maori and Europeans, 1773–1815. Auckland: Viking. Simmons, David. 1985. Whakairo: Maori Tribal Art. Oxford: Oxford University Press. Sissons, Jeffrey. 1998. ‘The Traditionalisation of the Maori Meeting House’, Oceania 69(1): 36–46. Smith, Linda Tuhiwai. 1999. Decolonizing Methodologies: Research and Indigenous Peoples. London and New York: Zed Books; Dunedin: University of Otago Press. Sorrenson, Maurice P.K. 1992. Manifest Duty: The Polynesian Society over 100 Years. Auckland: Polynesian Society. Sully, Dean (ed.). 2007. Decolonizing Conservation: Caring for Maori Meeting Houses Outside New Zealand. Walnut Creek: Left Coast Press Inc. Sundt, Richard. 2010. Whare Karakia: Maori Church Building, Decoration and Ritual in Aotearoa New Zealand, 1834–1863. Auckland: Auckland University Press. Tipa, Moana. 2005. ‘An Interview with Deidre Brown’, CS News. Chrysalis Seed Trust, 9 November. Waitangi Tribunal. 1999. Ngati Awa Raupatu Report. Wellington: Waitangi Tribunal, . 2004. Turanga Tangata Turanga Whenua: The Report on the Turanganui a Kiwa Claims. Wellington: Waitangi Tribunal.

2

The Junction of the Tala and the Itu ♦l♦

Athol Greentree

Introduction

P

acific architecture is often characterized by notions such as openness, exposed beams, pole construction, vibrant colours and lightness, to name a few. However, limiting the scope of Polynesian architecture to such labels is problematic. When architects and engineers are commissioned to design and build fale (classical Samoan houses), emphasis is given to producing the shape of a fale using Western ‘structural logic’ (McKay 2005: 33) and modern building codes. Recent examples include the Fale Pasifika at the University of Auckland, the National University of Sāmoa in Apia, the E.F.K.S Museum and Art Gallery in Sāmoa, and the new Sāmoa government offices in Māngere, Auckland. But what of Indigenous structural logic and architectonics? One layer of Pacific architecture often overlooked is the tectonic composition, necessary for the structural operation of this type of architecture, within the context of traditional buildings in the Pacific. Often the construction methods used are considered too primitive and structurally weak to be repeated.1 This chapter investigates the tectonic layer beneath the shell of the Samoan fale, paying particular attention to the junction of the round-ended tala with the gable-like itu section of the roof. This junction is critical to the architecture, enabling mobility and ensuring a structurally sound connection between two fundamentally different geometric forms. Architecture in



The Junction of the Tala and the Itu  25

Sāmoa, including construction detailing, is an established tradition, and as with any tradition of significance in Sāmoa, it has a citable genealogy with a mafua‘aga, or origin.

Figure 2.1. Fale interior and construction of a tala (rounded end). Samoan Material Culture, New Zealand Electronic Text Collection, Victoria University of Wellington. © Te Rangi Hiroa. CC A-SA 3.0.

26  Athol Greentree



Mafua’aga: Origins The oval roof form of the fale consists of three distinct parts: an itu gable roof span (the length of which varies) and two tala – that is, rounded ends that almost appear to have been constructed by extruding the sectional profile of the itu along a semi-circular path equal to half the itu width in radius. Both the appearance of the fale and its construction are unique to Sāmoa and Pacific Islands that have traditionally been within Sāmoa’s sphere of influence (Tonga, Niue, Uvea, Futuna, Rotuma). On the reasoning behind this method of closing off the gable ends, Mike Austin and Jeremy Treadwell (2009: 38) point out that priorities other than functional determinism are at play, as there are easier ways of construction – as evident elsewhere in the Pacific. These priorities can be traced back to a point of reference, a mafua‘aga. Tuatagaloa from Falealili recounts the history of the tufuga-faufale: In the time of Tagaloalagi, houses in Samoa were of different sizes and shapes and this led to much confusion and created many difficulties for those who wished to have a house erected as each carpenter was proficient in the building of one shape only. A fono of all the carpenters was held for the purpose of coming to a satisfactory decision relative to house building. Each carpenter wished that his style of house should be the one decided upon and the argument waxed enthusiastic. As there seemed to be no prospect of the matter being amicably decided Tagaloalagi was asked to decide. He pointed to the dome of heaven and the horizon and decreed that the shape of all houses should be as was the shape of the heavens, and it has been thus ever since. (Tuvale 1918: 6)

That Tagaloalagi’s mantra, that form should trace the outlines of the heavens, is the mafua‘aga or origin for all fale becomes evident when one observes the tala section of the house from the interior. The curved fau (purlins) which rise incrementally between the horizontal faulalo (curb plate) and the ‘au‘au (ridge beam) to mimic the rising of celestial bodies from the horizon to their highest point in the sky (Austin and Treadwell 2009: 39). It would be safe to assume that the craftsmen present at Tagaloalagi’s council were well acquainted with the construction of the gable, a typical method of construction elsewhere in the Pacific. Once curved purlins or fau had been chosen as the ‘bones’ of the tala, only one problem remains – how to solve the junction between the two geometrically different components. The problem required some negotiation between the tufuga-faufale clans concerning how to resolve the constructional sequence of building elements. A recent version of the oral history of this problem2 recounts the



The Junction of the Tala and the Itu  27

unresolved dispute between the numerous tufuga involved, the result being that the tala and itu were to be constructed as two separate elements. One faction built the tala while another commenced work on the itu. When the tala was completed and carried over to the itu for fitting, a mismatch was discovered – the gable edge/end of the tala did not meet the profile of the gable end of the itu. A solution was suggested by one tufuga: ‘Ia ta‘amilo pea ma tautala’ – meaning, ‘turn (the elements) and detach (the lashings) until a fit is reached’. A fit was achieved by twisting and manoeuvring the two components causing the faulalo (curved plate) ends to fa‘ase‘e ‘i lalo o le amopou (skew below the straight top plate). The ‘aso (battens) of the tala were pushed inwards so that they ‘tucked’ beneath the inner surface of the fatuga (edge battens) of the itu. In this way, the tala was only joined to the itu by the fatu-le-ulu-‘aso (top-end lashing of the ‘aso), and not by any kind of carpentry, giving rise to another proverb, ‘Ua vaea i ulu fatuga’ (divided at the fatuga). The methods employed by tufuga thenceforth were often referred to as togafiti (trickery), as the layperson did not readily understand them. Austin and Treadwell (2009: 39) suggest that the fale still maintains a ‘commitment to mobility’, indicating that Samoans are generally a mobile people and trend to frequently shift residences – for example, between village house site and plantation, paternal, maternal village and spouse’s village. When the need arose the lashings of the fale, along with the poulalo lashings, could be quickly undone effectively, separating the house into three sections and allowing each portion to be easily lifted and carried away by a group of men. This mobile design allowed for ‘shifting houses’ without damage to the parts, and once the parts arrived at the new location,

Figure 2.2. Typical fale sections. A, B and C depict three main points of the junction; A = moamoa; B = ulu‘aso and fatuga; and C = amopou and faulalo. © Athol Greentree.

28  Athol Greentree



they functioned as prefabricated components allowing for ease of assembly (Krämer 1995: 270). One documented case of ‘shifting houses’ is that of the faleafolau (larger meeting house with a longer itu) of Matai‘a in Vaimoso. The building was initially constructed as a faletele (short length of itu in comparison to faleafolau) on the seaside of the main road, where it was used extensively as a meeting place for the Mau movement during the 1920s. Later, the faletele was dismantled and shifted to the opposite side of the road, where the length of its itu was extended, making it a faleafolau (Van der Ryn 2012: 150). Many photographs taken during the period of civil war in Sāmoa, at the turn of the nineteenth century, depict the use of tala as shelters for gun batteries and checkpoints. The discussion will now turn to an examination of the technology facilitating the success of the junction. To fully comprehend its make-up, it is necessary to explain the three tectonic layers that comprise the form. The first and outermost layer consists of flexible timber battens (fatuga and ‘aso), tensioned over a system of transversely placed underpurlins (la‘au matua and fau) forming the second layer and, finally, a structural layer which braces and vertically supports the middle layer. The structural layers of the roof are named as follows: the fatuga (batten layer); the taotao (hold down layer); and the te‘e (propping layer). The itu section contains the te‘e layer that supports the ridge and pushes and pulls the taotao layer into position. The taotao layer consists of straight la‘au matua underpurlins in the itu and curved fau arches in the tala. The fatuga layer, comprised of fatuga and ‘aso, is then manipulated and lashed to the taotao layer resulting in the completed form.

Faulalo to the Amopou The faulalo (curved top plate) is the lowest element within the fau layer of the tala and is the first of all the fau to be fabricated. The faulalo is generally 70–100mm in diameter if constructed from a solid section of timber, or 70–100mm wide if from lumber. Unlike the other fau, it has a radius equal to approximately half the width of the itu and is the only fau that extends beyond the fatuga (boundary batten) to be joined to the itu. How this junction was dealt with in the mafua‘aga story is clear: it was thfa‘ase‘e (thrust or skewed) under the end of the amopou as a remedial measure. Fa‘ase‘e also means to embellish or flatter and is an important tool in any Samoan negotiation. What is intriguing about this junction is that although the tala is always tailor-made to suit the profile of the itu, the detailing in construction appears to re-enact the legendary or historical negotiations: ‘The curb plate is carried



The Junction of the Tala and the Itu  29

into position and temporarily strutted while its ends are lashed to the end rafters immediately below the wall plates. Some curb plates are cut away at the upper edge so as to fit against the wall plate, and bring the upper edges closer together’ (Buck 1930: 51). The lack of sophistication in the detailing of this junction appears to be deliberate. The seamless joinery of the fau and other similarly joined components attest to the tufuga’s skills, since joining the two elements to produce a singular and constant level ring beam would not have presented any difficulties. A variation on the method described by Te Rangi Hiroa (Sir Peter Henry Buck) was for the faulalo to be level with the amopou up until approximately two feet from the junction at which point the si‘ufau (final piece of the faulalo) skewed beneath the amopou. The hooked joint was sufficiently angled to accommodate the downward direction. My own family’s faletele was of this type. Only two other examples were personally observed in Savai‘i in 1995 and 2001. Only the persistence of tradition can explain the perseverance of this type of construction (one without fixings or joinery). The mafua‘aga was treated as a regulatory document for traditional building practice, the closest thing to a Samoan building code. Most new and existing traditional fale in Sāmoa today no longer demarcate the two elements as noted above; instead, they are joined together to form a single ring beam. Ring beams of this nature are known to have been built much earlier in the twentieth century – the N.Z. Centennial Exhibition fale built in 1939 in Sāmoa and shipped to New Zealand is one example. A ring beam may have been considered necessary for the Centennial Exhibition fale as it was to be erected on the exhibition floor without

Figure 2.3: In Buck (1930), Faulalo to Amopou junction – internal (b) and external (c) views. © Samoan Material Culture, New Zealand Electronic Text Collection, Victoria University of Wellington. CC A-SA 3.0.

30  Athol Greentree



cantilevered supports (Western Samoa Mail 1939: 5). The ring beam would have formed a structural element tying the tops of the posts together to provide stability. The move was also a possible attempt to portray Samoan architecture as civilized while on stage in a predominantly European setting, where a skewed faulalo might have been judged as primitive. An inspection of the eaves detailing in present-day faletalimalo (guesthouses) reveals a further rationale for the predominance of ring beam construction. Since at least the 1970s, Samoans have added tulutulu‘apa (timber-framed eaves) to enlarge the fa‘atautau (overhang) of the traditional roof. A constant level ring beam provides the necessary bearing and fixing support for the eaves framing, ensuring a uniform pitch and lip profile around the perimeter of the roof. In Tonga, where fale have walls, ring beams have been adopted to better accommodate new wall construction techniques. Timber studs and masonry blockwork walls, for instance, have been adapted to support the ring beam from the ground.

‘Au‘au to Moamoa Once the faulalo has been erected, the next stage is to create a transitionary element at the apex of the roof to join the centre ‘aso with the ridge beam using a peculiar element termed moamoa, which may take several forms. Buck (1930: 51) lists three while a fourth is described by Krämer (1995: 281).

Figure 2.4. The construction detailing of eaves in metal clad roofs necessitates a constant level ring beam. © Athol Greentree.



The Junction of the Tala and the Itu  31

1) A flat square or round piece lashed to a rebate in the bottom face of the ‘au‘au (ridge beam): A flat piece of wood barely as wide as the ridgepole, and from 7 to 9 inches long, was let into the under surface of the ridgepole which was cut… [the] under surface was also carved with curved lines as shown. Some are painted with dots and triangles, and others, according to Handy (14: 8), ‘are carved in symbolic representation of the moon and stars.’ From above the ridgepole overlaps part of the moamoa, and the even continuous slope downwards, by breaking the abrupt ending of a squared end, affords an even slope conforming to the plane of the middle rafters. (Buck 1930: 51)

2) A five-sided curved arch: ‘The construction of the second type is totally different. A curved piece of wood is placed transversely against the squared end of the ridgepole, and the outer edges of the end rafters, but is not lashed against them’ (Buck 1930: 51). 3) The uppermost fau, the fau tali‘aso (‘aso: receiver) is brought up higher so that it makes contact with the edge of the ridge: ‘Another variety was round in section, like a curved purlin, and extended directly upwards along the edges of the end rafters, its ends coinciding with the highest main purlins on either side of the middle section of the house. The curve is much sharper, and gives the lateral curve to the middle series of thatch rafters’ (Buck 1930: 51). 4) Krämer’s description of the fourth method bears some resemblance to (1) as also documented by Buck (1930: 50–51): ‘The centremost spar of the round part, the ‘aso o le totonu ends at its top in a ridge blade, a shovel like wider area called moamoa which is slid under a bay in the large ridge beam, whose projecting part is fittingly called fululupe “dove’s tail”’. That name is reminiscent of our dove-tailed rebating (Krämer 1995: 281). Of all the moamoa methods described above, only the first is fixed to the ridge beam, while in the other methods the moamoa comes in contact with edge members of itu but without any fixing. Buck (1930) gives no further details as to how the moamoa interacts with the ‘aso and whether they were connected or not. However, if the same principles are employed in the other three methods, i.e. a ‘meeting’ rather than a connection, it becomes apparent that the grooved underside could easily have provided housing for the ‘aso. In moamoa forms (1) and (4), the tops of the ‘aso are pinned down, creating tension as they are progressively lashed to the descending fau. In (2) and (3) the ‘aso make no connection with the ‘au‘au but instead ti‘eti‘e, or ride freely in place, so that the moamoa depends on central fatuga battens

32  Athol Greentree



or ‘aso rods already lashed to the curved purlins (fau) to keep it in place. As more ‘aso are added progressively in the lateral direction on either side, its position becomes more secure. The primary purpose of the moamoa in each case is consistently to regulate the upper ends of the ‘aso that are extended between the ridge and the faulalo. Without a moamoa of one form or another, the danger is that the ‘aso will overlap or extend too far beyond the edge fatuga of the itu causing an uneven surface that once thatched will affect the weather tightness of the building. Only the remaining three methods were observed in faletalimalo while only (3) has been used personally in construction. Moamoa (3) is by far the most common form in present-day fale while moamoa (2) is rare and appears to be a stylized variation of (3). The fululupe method was seen in use in many fale in both Sāmoa and Tonga – notable examples include the Sāmoa College faletele, the Tonga Visitor’s Bureau fale and the Lotofoa Wesleyan church (now demolished). Models of faletele in Stuttgart and the British museum also clearly show the use of the fululupe. These centrally located ‘aso, sometimes numbering four or five, are termed ‘aso o le totonu (Krämer 1995: 270–81), ‘aso vao (Buck 1930: 50–52) or ‘aso tau (Krämer 1995: 270–81). Alternatively, a single or double fatuga batten serves as a substitute but still maintains the terms described above. A further term associated with these ‘aso tau is mutiagiagi which refers to the deceitful nature of their connection with the ‘au‘au. From within the fale gazing upwards, they appear to be fastened in some complicated fashion – it is only during construction that one will realize that there is no fastening. They remain rigid only by their lashing to the moamoa.

Ulu‘aso to Fatuga With the centre ‘aso in place and securely lashed to the faulalo, fautu and moamoa, more ‘aso are added laterally to either side continuing down to the faulalo–amopou junction. The fixing method of the ‘aso at their upper ends (ulu‘aso) switches from over to under as they depart from the moamoa. Again, the method employed references the mafua‘aga story where the ulu‘aso were pushed and manipulated inwards in order to tuck them under the edges of the gable. In plain view, the ‘aso in the tala are placed at right angles to the fatuga and ‘aso of the itu. The ‘aso are lashed flat to the underside of the fatuga (of the itu section), a method called fatu (to compose) creating tension as each one is pulled down and lashed to each of the fau, progressing downwards from above. In this way, the only real connection between the tala and the itu is the lashing of the ulu‘aso, which, as pointed out by Austin and Treadwell (2009: 8),



The Junction of the Tala and the Itu  33

resembles ‘stitching’. Buck asserts that this junction is a weak point of the architecture: ‘The weakness of Samoan houses is the joint of the rounded ends to the end rafters of the middle section. There is danger that the wind will lift the thatch directly and take the roof with it’ (Buck 1930: 82). This concern, however, is perhaps a little unfounded. In the major cyclones over the past 100 years, cyclone damage to fale has been limited to failure of the vertical supports, the bases of which have been insufficiently sunken into the ground or have succumbed to rot (Lapish and Lynch 1992: 5–20). A more common consequence was for the whole roof to be lifted upwards. Cases where the tala separated from the itu have been the result of collisions with substantial debris such as cars; an example was witnessed in the 2009 Samoan tsunami. Wind damage to the tala–itu junction is minimal.

Figure 2.5. Ulu‘aso lashed to the underside of the fatuga batten. © Athol Greentree.

34  Athol Greentree



Often, an additional ‘aso is lashed transversely to the inner surface of the ulu‘aso from the moamoa continuing downwards toward the amopou– faulalo junction, running parallel to the boundary fatuga. At each ‘aso that it crosses over, a lashing is applied. This serves to strengthen and brace the ‘aso to resist lateral movement. If the tala is to be detached and moved, this vertical ‘aso, like the role of the moamoa, prevents disfiguration of the gable-like profile. The same method was practised in Tongan fale, where there is no distinction in size between fatuga and ‘aso – all are uniform throughout, being roughly 50 by 25mm and termed kahoki. Two smaller domestic fale in Ha‘apai and one in Tongatapu exhibited additional strengthening at the junction in the form of short battens of the same dimension as the kahoki but roughly 150mm in length. These were placed horizontally over the edge kahoki of the itu and the vertical kahoki in the tala at approximately 100 to 150mm spacing. The finished effect was rather pleasing and resembled a continuous cross stitch from the ring beam to the ridge. A potential weakness in the architecture was resolved using simple lashing at close centres. Buck (1930: 53) writes that: ‘The stability of the house shows the strength of a combination of weak elements, such as the thatch rafters’. In contrast to the faletalimalo, the faleo‘o, on the other hand, lacks the careful bridging of the tala and itu. The ‘aso are instead lashed to the iviivi (straight solid timber rafters). as opposed to the boundary fatuga which have been curved with the use of ‘graduated chocks’ (Buck 1930: 18). The result is a gaping hole at the junction – once this is thatched over, the demarcation between the three sections is intensified.

Conclusion The junction of the tala and itu is a forgotten and obscure but vital principal element of Samoan architecture. This chapter reveals that the fale consists of two tectonically unique geometric forms making a total of three roofs – one itu and two tala. This is in contrast to modern representations of fale that through structural logic have come to consist of a single homogenous shell where the tala becomes an extension of the itu, both structurally and aesthetically. The construction detailing employed in the fale recalls a commitment to mobility forged in ancient inter-guild negotiations between the tufuga-faufale. The momentous inter-guild negotiations became mafua‘aga, a reference point and the equivalent of an acceptable solution that is used by architects and engineers. The junction follows two fundamental principles. First, the elements at the apex and lowest points should ‘meet’ rather than be ‘joined’. This is manifest at the apex where the mutiagiagi bluff a



The Junction of the Tala and the Itu  35

connection while no such effort is made at the faulalo. Secondly, structural strength to resist imposed loads at the junction is achieved through employing numerous weak elements consisting of ‘aso and sennit lashings at close spacing echoing the Samoan proverb ‘Tasi ‘ae afe’ (One, but a thousand). The strength of these weak elements and their resistance to imposed structural loads is dependent on binding technology. It is difficult to realize this type of construction with nail fixings because ‘aso rafters are often of a size where they can be easily split by nail. Pacific spaces and buildings, from the perspective of this chapter, employ methods more often associated with poetry and song than construction – fatu (to compose), fa‘ase‘e (to embellish), ti‘eti‘e (to ride) and togafiti (trickery). Aesthetic qualities arise from functional necessities alone rather than ornament. The orchestration of tectonic elements combines to achieve maximum structural performance resulting in a state of mālie (satisfaction) and matagofie (beauty). In this light, fale arising from the modern-day construction industry are often merely code-complying palagi (Western) structures with a token layer of added Pacificness where bolts are concealed with lashing, and motifs – both painted and carved – are placed on the exterior and interior. While it is understood that institutions of the fa’a Sāmoa (Samoan way) such as the fono (meetings) can and do function without tufuga-built fale, they are still sought after and held in high esteem both in Sāmoa and in the diaspora. Architects, engineers and builders, therefore, will continue to assume control of fale production. An understanding of mafua‘aga and their manifestation in construction, as discussed in this chapter, would result in more interesting architectural developments, one where Samoaness does not rely on ornaments or decoration. Athol Greentree is a practising architect working in Sāmoa, and a graduate of the Master of Architecture (Professional) course at the University of Auckland. His thesis is entitled ‘Samoan Tectonics: A Study in the Tectonic Operation of the Samoan Fale’, and his research interests include architecture and landscape architecture, archaeology and heritage conservation, sustainable building design and prefabrication.

36  Athol Greentree



Notes   1. The most active of traditional fale builders in Sāmoa, Fonoti Likisone, relayed that he believed traditional Samoan tufuga-faufale (housebuilders) built inferior buildings in comparison to modern architecture, due to tufuga-faufale not possessing enough structural knowledge.   2. From Notoa Faatafea, Fiu Ga‘utavai and Manase Apelu (personal communication).

References Austin, Mike and Jeremy Treadwell. 2009. ‘Constructing the Pacific Hut’, Interstices: Adam’s House in the Pacific 10: 21–31. Buck, Peter. 1930. Samoan Material Culture (Vol. Bulletin 75). Honolulu: Bernice P. Bishop Museum. Handy Edward Smith and Willowdean C. Handy. 1924. Samoan House Building, Cooking, and Tattooing. Honolulu: Bernice P. Bishop Museum. Krämer, Augustin. 1995. The Samoan Islands: Vol. 2 (trans. Theodore Verhaaren). Auckland: Polynesian Press. Lapish, Ernest B. and Alan John Lynch. 1992. Strengthening Village Fales for Strong Winds = Faamalosia o Fale o Alalafaga Ta’itasi Puipuia ai mai Matagi Malolosi. Public Works Department, Government of Western Samoa in conjunction with the Ministry of External Relations and Trade, Government of New Zealand. Wellington: The Ministry. McKay, Bill. 2005. ‘Poly-technic’, Architecture New Zealand, May. Refiti, Albert. 2010. ‘Whiteness, Smoothing and the Origin of Samoan Architecture’, Interstices: Adam’s House in the Pacific 10: 9–19. Simmons, Lynda. 2011. ‘Interior Darkness / Contained Shadow’, Interstices: Unsettled Containers 12: 126–30. Treadwell, Jeremy. 2005a. ‘Chains of Negotiations: Navigating between Modernity and Tradition’, Interstices 6: 110–15. . 2005b. ‘Post Colonial Fabrication’, Architecture New Zealand, May. Tuvale, Te‘o. 1918. ‘The Samoan House – O le Pale Samoa’, An Account of Samoan History up to 1918. Tidal Pools: Digitized Texts from Oceania for Samoan and Pacific Studies. Van der Ryn, Micah. 2012. ‘The Difference Walls Make’, PhD thesis. Auckland: University of Auckland. Western Samoa Mail, 25 November 1939.

3

The Energetics of Vā and the Samoan Faletele ♦l♦

I‘uogafa Tuagalu

Introduction

T

he Samoan vā, so central to Samoan worldviews, is often described as relational space. This chapter examines the ontology of vā, focusing on what people believe vā to be. The notion of Samoan vā (between-space) is explored in terms of a field of forces, a vā-field, and applied to Samoan artefacts: the faletele (round meeting house) and faleafolau (large oval guest or meeting house). While Samoan spatial concepts differ in important ways from Western traditions, there are some conceptual touching points. For instance, in thinking about space, Western traditions rely on Aristotles’ notion of a ‘nested series of places, up to the outer sphere containing the universe’ (Levinson 2003). They can be categorized as, for instance, mathematical space, objective space, social space or psychological space, which each entail particular ontological commitments for entities in those spaces. In a typology of the English term ‘space’, I have shown how objective physical and topological spaces are base notions of space, describing length, height, depth, proximity and adjacency (Tuagalu 2015). These spatial characteristics are then applied metaphorically to other types of space. Pierre Bourdieu, for example, refers to the discipline of sociology as ‘a

38  I ‘uogafa Tuagalu



social topology’, where ‘the social world can be represented as a space (with several dimensions) constructed on the basis of principles of differentiation or distribution constituted by the set of properties active within the social universe in question … Agents and groups of agents are thus defined by their relative positions within that space’ (1985: 723–24). Spaces, according to Bourdieu, are thus conceived in sociology as social constructions and relational (topological) – a point of convergence with Samoan perception of vā.

The Samoan Vā The two basic dictionary meanings of the term vā that remain uniform over 150 years of recorded Samoan lexicography are, as a noun, ‘a space between’ (resonating with Western topological and objective space) and, as a verb, ‘to rival’. However, there is a discernible shift of emphasis in the usage of the term, away from vā with objective and topological connotations to vā as social relation. The earliest Samoan–English dictionary was compiled in 1862 by George Pratt, one of the first European missionaries in Sāmoa, who involved recently converted Samoan experts in compiling his dictionary. In that dictionary, the two meanings of the term vā are first delineated and defined: firstly, as ‘a space between’, or ‘to have a space between’;1 and, secondly, as a verb with the meaning ‘to rival’ (see Pratt 1862: 216; see also 1893: 331). In 1966, George Bertram Milner (1918–2012) published his Samoan Dictionary, updating the Pratt dictionaries and highlighting changes in usage. Here, Milner’s entry for the term vā is extended. The main meaning of vā is still ‘distance, space (between two places, things or people)’; but there is an emphasis on the social meaning of ‘relationship, relations (between two things or people)’ (Milner 2003: 307). Papaāli‘i Dr Semisi Ma‘ia‘i published a two-volume Samoan-English dictionary, titled Tusi‘upu Sāmoa, in 2010. The term vā has the base meaning of ‘space, interval’; and, as a prefix, ‘creating space’ (Papaāli‘i 2010: 453). Papaāli‘i also emphasizes the definition of vā as rivalry or to rival. For Papaāli‘i, rivalry and contestation are a central feature of Samoan vā relations. He translates the Samoan proverb ‘se‘i teu le vā’ as ‘let the variance of opinions be modified’ (Papaāli‘i 2010: 453).

The Vā in the Diaspora In 1996, Albert Wendt, the renowned Samoan writer, had first articulated in written English the importance of the Samoan vā as relational-social space:



The Energetics of Vā and the Samoan Faletele  39 Important to the Samoan view of reality is the concept of Va … Va is the space between, the betweenness, not empty space, not space that separates but space that relates, that holds separate entities and things together in the Unity-that-is-All, the space that is context, giving meaning to things. The meanings change as the relationships/the contexts change … A well-kown [sic] Samoan expression is ‘Ia teu le va.’ Cherish/nurse/care for the vā, the relationships. This is crucial in communal cultures that value group, unity, more than individualism: who perceive the individual person/creature/thing in terms of group, in terms of vā, relationships. (Wendt 1996)

Wendt redefines the traditional use of the term vā.2 In terms of Samoan grammar, ‘vā’ is an instance of what linguists Mosel and Hovdhaugen (1992: 94) call ‘local nouns’, i.e. nouns that denote the ‘spatial dimensions of objects and directions’.3 The objects whose relation are denoted by a local noun are usually stated. So, in terms of vā, the objects in vā relations would be itemized, e.g. vā o ali‘i ma tulafale (the relationships between chiefs and orators), vā o mātua ma fanau (relationship between parents and children). Local nouns denote a topological (i.e. non-metric) relation between objects. However, Wendt uses vā as a stand-alone abstract noun to express the sense of relatedness or relational space, which he terms ‘betweenness’. Since then, Wendt’s often quoted description of vā has become the touchstone for diasporic4 Pacific artists and scholars striving to express a unique Pacific worldview based on home-island understandings of vā, but transposed to the new homelands to which Pacific peoples had migrated. The concept vā is now equated with relational space and often referred to as ‘relationships’.5 Quite often, diasporic explications of vā are inextricably intertwined with ideas of Pacific identity, that is, they render vā-relatedness as unique to Pacific peoples, forming Pacific communities in the new countries to which they have migrated. Samoan vā tend to include human agents, and can include non-human and inanimate ones (Tui Atua 2009b: 116).6 Furthermore, I have argued elsewhere that the different types of vā also tend to fall between two interrelated types: vā fealoaloa‘i (social space) and vā tapua‘i (sacred/spiritual space) (Tuagalu 2008: 111).

Vā Fealoaloa‘i (Social Space) For Samoans, social vā is about knowing one’s position in relation to others. Island Samoans learn their sense of place in a village context, through interactions with village institutions. Members of each division of Samoan society7 have social responsibilities towards, and communication protocols with, members of each of the other social divisions. Samoans learn social

40  I ‘uogafa Tuagalu



vā by interacting with their peers, superiors and inferiors. The importance of social vā, family status and ranking, can be seen in the fa‘alupega of the village (a charter of the founding titles of a village). Every village member and their family can place themselves within the fa‘alupega. It is ‘a verbal distillate of history and a who’s who of a community, a district and even of all Samoa’ (Keesing and Keesing 1973: 74).8 Diasporic Samoans do not live in traditional Samoan villages in their new homelands. Those who wish to continue practising homeland cultural forms organize themselves around other social institutions, such as churches. In these institutions, they organize themselves along chiefly and family groupings and practise social gift-giving rituals, e.g. the distribution of ‘ie toga (fine mats) at weddings and funerals and ‘ava (kava) rituals at chiefly meetings.

Vā Tapua‘i (Sacred Space) In the nineteenth century, tapua‘i had the meaning ‘to abstain from all work … and to sit waiting for success in war or in sickness … To give something to bring success’ (Pratt 1893: 303).9 The modern meaning of the term tapua‘i is to ‘sanctify, make holy’, or the moral support of a chief ’s lauga, oration (Papaāli‘i 2010: 389). Fanaafi Aiono Le Tagaloa (2003: 49) offers a contemporary translation of the term tapua‘i, as ‘the worship of God for the good of someone else’. An example of the vā tapua‘i in action is the situation where people remaining at home (the au tapua‘i, meaning literally ‘those who wait and pray’) would pray for the successful outcome and safe return of a travelling party (the au malaga). These prayers in support are considered essential for success. In other words, even though they are geographically distant, the travellers and supporters are in what might be called ‘contagious magical contact’.10 A defining characteristic of the Samoan vā is ‘action at a distance’, i.e. ‘a direct and instantaneous interaction between bodies that are not in physical contact with each other’ (‘Action at a Distance’, 2015). Vā tapua‘i has to be activated via prayer or ritual. Vā fealoaloa‘i (social space) and vā tapua‘i (sacred space) are not opposite ends of a continuum; rather, they are closely intertwined. For every socially defined vā, there is an underlying vā tapua‘i. For every vā tapua‘i, there are corresponding social relations determining whom and what it affects, as well as appropriate social rituals for its proper engagement. The paradigm instance of vā is the feagaiga (covenant relationship) between brother and sister. Socially, kinship entails a special set of reciprocal obligations: in this case, the brother serves his sister in terms of resources and, in turn, the sister acts as advisor to her brother; she tapua‘i (supports, prays for the success of) all his endeavours.11 The feagaiga illustrates the



The Energetics of Vā and the Samoan Faletele  41

ideal instance of kin relations in the social vā, as well as the activated spiritual linkages of the underlying spiritual vā (vā tapua‘i) of the feagaiga. All vā have to be regularly activated and maintained to remain viable. The Samoan proverb ‘ia teu le vā’ (tend to the vā) cautions that vā needs tending, much like a garden. When vā fealoaloa‘i are not tended properly, the vā attenuates and is ‘turned off ’, or negative social vā or relations arise and social discord will increase. When vā tapua‘i are not nurtured appropriately, then the opposite of tapua‘i (sacredness or support) can occur, namely, fetu‘u or malaia (curses).

Forces and Fields To recapitulate: the basic notion of vā is the distance between objects, and though these objects may be spatially separated, they are connected in a vā relation. What, then, is the nature of this connection? I will explore this connection here through the lens of forces and fields. In the physical sciences, a field is defined as ‘a region in which a body experiences a force as the result of the presence of some other body or bodies’ (‘Field’, 2015), like in a magnetic field generated by a magnet attracting an iron object. The concept of field, that is, a region under the influence of some force, offers an explanation of ‘action at a distance’ – in that objects that seem spatially separate are joined by forces within a field (see Allen 2005; Hesse 2005). Other writers12 have used the notion of field to examine Pacific spatiality (Bennardo 2009; Lehman and Herdrich 2002) in particular. F.K. Lehman and David Herdrich use the notion of point field to examine Samoan spatiality: ‘a point field defines space as the topological neighbourhood of a given point, and boundaries are derived as the adjacency of the closures of pairwise distinct point fields’ (181; emphasis in original). In their conception, a point, or object, generates an infinite field, which forms a boundary when its field runs up against another point field. They contrast point-field space with absolute space – space being formed by two points in relation to one another (relational space), as opposed to space existing independently. Giovanni Bennardo (2009: 173) discerns a principal radiality in Tongan use of spatial terms: ‘radiality is the relationship between two points, where one of them functions as the origin or goal of the vector that signals the relationship’. For Bennardo, this radial relationship is foundational to Tongan ways of thinking, in that movement towards a centre is denoted by the Tongan term mai, and away from centre, atu. This is a basic organizational structure, a ‘cognitive molecule’. This language pattern (of movement either towards or away from the speaker or other subject) can be found repeating itself throughout all Tongan knowledge domains, e.g. kinship, social

42  I ‘uogafa Tuagalu



hierarchy and gift-giving. Bennardo notes a Tongan linguistic preference for expressing spatial relations in terms of ‘the “radial” subtype of the absolute FoR [Frame of Reference]. That is, a specific point of reference is chosen in the environment of the speaker, and then spatial relationships are expressed as toward or away-from it’ (Bennardo 2009: 75). Lehman and Herdrich, as well as Bennardo, emphasize the location, or orientation, of the various points in relation to each other. The mathematical term ‘vector’ describes these spatial relations and orientations to some extent, but does not address the role of the forces that either attract or repel the objects within that space or region.13 In his analysis of Tongan spatiality, Bennardo (2009: 188) mentions the supernatural force mana, but only to illustrate ‘that radiality – in this case, radiating power from other-than ego’ – is fundamental to traditional Tongan religion. Despite Christianity replacing previous Tongan religious belief systems, radiality is still the ‘fundamental way of conceptualizing the religious domain’; it motivates the way in which people behave ‘at least conceptually’ (Bennardo 2009: 189). However, forces within the field both define and limit the extent of the field, and different spaces may be subject to different forces. For example, social forces would control social space; psychological forces would shape psychological space; physical forces generate physical/objective space; and topological vector spaces would be oriented by direction and magnitude and amenable to mathematical operations. Vā can be conceived as a field, in which objects are located and wherein vā-forces operate. Mana, tapu and alofa are Pacific concepts that have been examined at length (Shore 1989; Tomlinson and Tengan 2016), and they are forces that may bind or repel objects in vā relations. Mana serves to channel the influence of the gods towards generative ends; it is attributed to and can be accumulated in a person (Shore 1989: 140). Mana is appropriated, almost wrested from the gods. Tapu is a restrictive, inhibitive force, which can channel generative power. Both mana and tapu work in tandem. Negative instances of mana and tapu lead to a withering of generative power, to societal disorder and fissures in the cosmic order of the world. The Samoan concept of alofa, commonly used to mean love or compassion, is another such binding force. To present-day Samoans, alofa has become closely associated with the Christian notion of agape (Papaāli‘i 2010: 15) and, in this form, Christian Samoans speak of alofa14 as a force motivating most positive Samoan activity, in the form of manatu lelei (good thoughts) and socially positive actions. The opposite of alofa is the privative state of lē alofa (devoid of alofa). For example, a reverend who tends well to his flock would be spoken of as



The Energetics of Vā and the Samoan Faletele  43

having alofa for his parishioners. A reverend who embezzles funds from his parishioners would be accused of being lē alofa, having ‘no love’ for his parishioners. Interestingly, Shore (1996: 302) states that ‘alofa represents both a positive feeling of compassion and identification and a defence against hostile feelings commonly associated with deferential relations’. In Sāmoa, mana, tapu and alofa exist, in Western terms, both intramentally and extramentally: they are part of an objective world, and they have a realist ontology. That is, Samoans believe that these forces exist in the objective or real world. One can contend that, just as different types of space may be subject to different forces, forces like mana, tapu or alofa are differently configured in various types of vā.

The Faletele and Faleafolau I will now show how the notions of field and vā relate to significant Samoan artefacts such as the faletele and faleafolau (meeting house or guest house). The large Samoan fale (house) comes in two varieties: the faletele, which is characterized by a large, almost conical roof and round paepae (raised floor base), and the faleafolau which has a large, elliptical roof with a corresponding elliptical paepae. One can discern aspects of vā in these buidings in several ways. Any fale is, firstly, a physical structure that occupies (measurable) space. Its dimensions were traditionally measured in terms of gafa, the breadth of a man’s outstretched arms (approximately 6ft); half-gafa (outstretched arm to centre of chest), valuaga o le lima (fingertips to elbow ), and afa (outstretched hand from thumb to index) (UNESCO 1992: 18). As a physical structure, it is subject to load-bearing forces. For example, the roof is hung from the central poutū (main poles); the sides of the roof do not fall inwards because supporting ribs and lattice frames exert an outward force that helps to maintain the roof ’s convex shape. The two tala (ends of the house) are independent structures that are ‘clipped on’ once the main central structure (poutū and itū, the middle section of the roof) is in place. Finally, the outer pou (poles) support the edge of the roof. Secondly, fale exists in a social vā with the people of the village, functioning as a meeting house for the village council and/or as the domestic centre for the family of the matai (chief). Its existence is a status marker for the matai, as it testifies to their ability to organize the accumulation of resources for fale construction, including the rituals and gift-giving ceremonies associated with contracting, maintaining and paying the tufugafaufale (master builders), and adds to chiefly mana (Tuagalu 1988: 7). Thirdly, the fale is a centre for ritual. One of the most important rituals associated with the fale is the ‘ava ceremony – the imbibing of kava at the

44  I ‘uogafa Tuagalu



meetings of matai. This ritual reinforces the ranking of the matai, but it also links back to the deity Tagaloa and his alofa for humankind, a reminder of the divine vā tapua‘i. Fourthly, the faletele serves a religious function. One can see this in its origin stories in which Tagaloa, the supreme atua (deity), figures predominantly. In the creation song, Solo o le vā,15 Tagaloa assigns the specialist house-building duty to the guild of Sā-Tagaloa (Tagaloa family). As a result, people must seek the permission of the Tagaloa to build houses: in some versions of the Solo, Tagaloa comes down from the heavens and destroys the Tui Manua’s illicitly built fale. Furthermore, the large, dome structure of the fale roof, propped up by the central pou in the faletele, is said to mirror the Samoan cosmos. According to some Samoan creation myths,16 Tuite‘elagi (Chief of the heaven-‘proppers’), at the command of Tagaloa, used the broad-leafed Teve (Amorphophallus Paeoniifolius, commonly known as Elephant Foot Yam) and Masoā plants (Tacca Leontopetaloides, a type of Arrowroot)17 not only to separate and prop up the sky from the earth,18 but also to prise apart and prop up further openings within the firmament. These openings then became the ten distinct lagi (heavens) of the Samoan cosmos, which were populated, according to another creation account, by the offspring of earth and sky: Ilu (immensity) and Mamao (distant).19 Nine lagi are inhabited by the offspring of Ilu and Mamao, but the tenth lagi is for Tagaloa alone. In this way, the central pou of the fale is representative of the primordial Teve and Masoā plants, and the tiered rafters of the lagi. Activities in the fale would therefore take place underneath the Samoan cosmos.

Figure 3.1. Flows of mana energy from Lagi through the faletele and the assembled chefs to the village. © Albert L. Refiti.



The Energetics of Vā and the Samoan Faletele  45

Refiti (2014: 58) views the fale as a channel for mana (divine energies). From the heavens, mana flows down from the sloping roof, concentrates around the alofisā (sacred meeting circle) of chiefs, and then pulses outward through the malae (village centre) to the periphery of the village. The field containing mana is strongest at the centre and diminishes in power the further it moves away from the centre. Conversely, vā-energy is channelled from the earth heavenwards through the alofisā and the rituals taking place in the fale. The genealogies of ali‘i (high chiefs) trace back to the gods (Gilson 1970: 23–24; Meleisea 1987: 8), whereas the tulāfale (talking chiefs) speak and act on behalf of ali’i. For Refiti, the genealogical links to ancestors – which the chiefs, as the latest incumbents of their chiefly titles, represent – provide the dynamism for the flows of energy. Shore (1977: 444) notes that the authority practised by the ali‘i is better termed mana (sacred power), whereas the tulāfale exercise executive authority.20 In the faletele, ali‘i sit at opposite tala (ends) of the fale, facing one another, and tulāfale are arranged by rank closest to the ali‘i they serve. In terms of energy flows, ali‘i pulse mana and tapua‘i energy to tulāfale, who are then ‘powered up’ to activate the various divisions of the village to carry out the will of the ali‘i.21 This dynamic is acted out in every ritual involving Samoan chiefs. In summary, the Samoan fale is not only a physical structure functioning as a meeting place for the village chiefs and a centre for their domestic activities, but also a miniature representation of the Samoan cosmos22 and that primordial architectural act, the separation of the Sky Father from the Earth Mother.23

Conclusion In a vā relation, two objects may be spatially separated but still affect one another. One way of conceptualizing vā relations is through the notion of field, in which objects are subject to forces. Three forces, which have been widely reported to attract and repel objects in vā relations and to motivate Samoan behaviour, are mana, tapu and alofa. Mana is generative, and tapu is restrictive or inhibitive. Alofa is a socially positive, binding force. Differing configurations of mana, tapu and alofa give rise to different types of vā. The Samoan universe is held together by the channelling of these forces through ceremonies and activities of its matai, people and the environment. The significance of vā-forces and vā-fields is evident when one examines the vā-energy flowing around significant material artefacts, such as the faletele and faleafolau. Tagaloa, the primary deity, sanctioned the construction of fale by the specialist guild, Sā-Tagaloa. This divine origin imbues

46  I ‘uogafa Tuagalu



the faletele with considerable sacred power (vā tapua‘i). One can see the binding and repelling properties of the vā-forces through the chiefly rituals that take place under its roof. Every object within the vā-field is subject to mana, tapu and alofa forces. Chains of vā relations are sparked with each individual vā interaction, drawing in and generating vā-forces and creating their own vā-fields. Perhaps surprisingly, the meanings of the Samoan term vā, as ‘space between’ and ‘to rival’, have been lexicographically consistent for the last 150 years. Vā fealoaloa‘i (social vā) and vā tapua‘i (sacred vā) are essential to understanding what motivates Samoan behaviour in traditional island and village-based settings. Diasporic Samoans use the term vā as an abstract noun to mean relatedness. Removed from traditional island settings, Samoans in the diaspora have adapted vā as an identity marker, as a way of differentiating themselves from other communities, and carrying out those cultural forms they wish to retain.

Postscript The Samoan vā worldview has been in formation for at least a millennium; it changed during that time24 and is still subject to change. The energetic view of vā, its forces, fields and relations, enables us to track what is being changed, adapted and retained in Samoan worldviews. This points to a major challenge for modern homeland and diasporic Samoan communities: to find alternative energy sources to power the Samoan vā networks they wish to preserve or adapt. Both sets of communities face the same problems, namely the need to maintain or adapt a Samoan vā worldview in the face of increasing global, political and economic forces affecting changes to the energetics surrounding vā-forces, vā-fields and vā relations. Diasporic Samoan communities live and work in foreign lands; they cannot draw on the traditional village, regional, chiefly arrangements and family settings that propel vā-forces. They have adapted vā-fields in that diasporic church communities have replaced the traditional village setting as a focus to practise chiefly and social rituals, for example. For both homeland and diasporic communities, this process involves finding (or, perhaps, redefining) energy sources to power their social and sacred networks, and refining their sense of identity, as they make choices as to what aspects of their culture they will maintain. I‘uogafa Tuagalu explores how Pacific concepts indicate a particular worldview; and how those Pacific worldviews may augment, or offer differing views of, current interpretations of Pacific history. He has written and presented a number of papers on the notion of the Samoan vā, and



The Energetics of Vā and the Samoan Faletele  47

the utilization of this concept in the disciplines of Education and Spatial Design. Tuagalu currently works as a Learning Advisor for Te Mātāpuna Auckland University of Technology, Library.

Notes  1. Examples of this usage of vā, as an indication of distance or separation between two or more objects, can be seen in nineteenth-century Samoan literature: for example, ‘sa nofo i le vā o Afega ma Malie’ (they lived on the land between the villages of Afega and Malie; Steubel 1896: 170); in the Samoan Bible, ‘i le vā o vāi le vānimonimo, e vā a’i isi vāi ma isi vāi’ (Genesis 1:6, ‘Let there be a firmament in the midst of the waters, and let it divide the waters from the waters’); and in poetry, ‘ua ta moe i le vā o tama’ita’i?’ (‘or did I sleep [in the space] between two girls?’; Krämer 1994: 262).   2. Tui Atua states that the term tuāoi (boundary) and vā are interchangeable depending on who is speaking. Participants in a vā relationship would speak of the vā with the other. Someone commenting on or evaluating the vā of others would speak of tuāoi (see Tui Atua 2015: 6–7, on tuāoi (boundary) as a result of vā; Tui Atua 2009a: 161–66, for a discussion of tuāoi in reference to the Samoan Lands and Titles Court).   3. See Mosel and Hovdhaugen (1992: 94–97) for a list of local nouns.   4. The Pacific diaspora refers to the migrations of Pacific Islanders, who since the 1950s moved from Pacific homelands and settled in more developed countries such as New Zealand, Australia and the United States.   5. See Anae (2010), Ka‘ili (2005) and Mila-Schaaf and Hudson (n.d.) as representative of the diasporic approaches and uses of the concept of vā over a range of disciplines.   6. Tui Atua (2009a: 116) has written on vā tapuia, the ‘sacred (tapu-ia) relationship (va) between humans and all things, animate and inanimate that connects the entire cosmos’.  7. Aiono (1997: 6) notes five social divisions, or saofaiga (groupings), in a Samoan village: the matai (chiefs); faletua ma tausi (wives of matai); tama‘ita‘i (unmarried women); aumaga (untitled men) and fanau lalovāoa (children). Each of the divisions has protocols that govern the interaction of its members with those of other divisions.   8. Collections of fa‘alupega play a very important role in traditional Samoan social and ceremonial life. See Krämer (1994, originally published in German in 1902); Le Mamea et al. (1958); The Tusi Fa‘alupega Committee (1985); Aiono (1997); and Tofaeono (2012).   9. Pratt (1893: 303) notes that the meaning ‘to offer religious worship’ is a ‘recent adaptation of the word’. He lists the usual meanings as ‘1. To abstain from all work, games, &c., and to sit waiting for success in war or in sickness. 2.

48  I ‘uogafa Tuagalu

10.

11.

12.

13.

14.



Applied to passengers in a canoe thanking the pullers, who answer, Fa‘afetai tapua‘i!’. In modern usage, tapua‘i commonly means ‘to support’. Frazer (1967: 49) describes contagious magic (one of the two branches of sympathetic magic, the other branch being homoeopathic magic) as ‘the notion that things which have been conjoined must remain ever afterwards, even when quite dissevered from each other, in such a sympathetic relation that whatever is done to the one must similarly affect the other’. Accordingly, elements in the vā can be viewed as being bound together by contagious magical connections, whereby they can affect one another even though they may be spatially separated. The illustration of the feagaiga can be seen in the relationship of the mythical Leutogitupa‘itea and her brother Lafaitaulupo‘o. The story goes thus: Leutogi becomes a junior wife to the Tu‘i Tonga and kills the baby son of her husband to one of his principal wives. Leutogi is condemned to be burnt alive for that infanticide. On the day of the execution, her brother, Lafai, having sensed his sister’s distress, sends a swarm of bats (the family totem) from Sāmoa to Tonga, and they extinguish the flames of the pyre by urinating on them. The Tu‘i Tonga then banishes Leutogi to a desert island, where those same bats feed her with fruit. Leutogi is saved from the island by the Tuiuea, and her son returns to her brother’s people in Savai‘i with the major titles of Tonumaipe’a, Tauiliili, and Tilotilomai. The brother protects the sister and follows her directions; the sister serves her brother by providing advice. See Krämer (1994: 121–22) for a modern rendition of the myth. See also Lafai-Sauoaiga (2000: 67–70) . Bourdieu also applies the notion of field to his theory of social fields. Hilgers and Mangez (2015: 1–40) trace the lineage of Bourdieu’s use of the term ‘field’ in the physical sciences, as well as in psychology through the Gestalt therapist Kurt Lewin. They describe the Bourdieusian field as ‘a structure of relative positions within which the actors and groups think, act and take positions’ (10). Fields are social constructs. Bradd Shore (1982: 186) notes, for instance, that Samoan behaviour is played out in two contrasting social contexts that the individual inhabits: the public and the private. Aga, public behaviour, is the ideal behaviour expected from Samoans under social constraints and guidelines. Amio, private behaviour, is composed of the personal feelings and true desires of the Samoan individual, unfettered of social constraints and taboos. Amio and aga motivations are constantly at odds with one another, as aga controls the private impulses of amio. See Afoa (2012: 27–30) for types of alofa in the Bible. Details about this aspect are beyond the scope of this chapter; suffice it to say that further research on tracing pre-Christian understandings of alofa is required. Afoa’s entry under akape (agape) is of interest: the commonly used gloss for God’s alofa (love). Afoa also links the term agape to the ‘lovefeast’ (66). Elsewhere, alofa seems linked to food production and consumption in Samoan myths; see Steubel



The Energetics of Vā and the Samoan Faletele  49

(1896: 170 [drinking kava]; 171 [cannibalism]; 173 [umu-traditional cooking method]). 15. See Powell (1887). 16. Both Krämer (1994: 539–44) and Fraser (1897) have slightly differing versions of the creation myth, Le Solo o le vā. 17. Krämer (1994: 540, 543, in footnotes 175, 176) notes that the leaves of the Masoā spread wide at the top so they would act as a platform to support the sky. Forcing people to eat the tuber of the Teve was a form of punishment (Steubel 1896: 135, 220). The Teve is also known as the ‘stink lily’ and is related to the Amorphophallus Titanum, whose inflorescence has the odour of rotting flesh that attracts flies. 18. In Māori mythology, it is Tane, the god of the forests, who separates his parents Rangi, the sky, and Papa, the earth, thus ending their perpetual coitus (Grey 1965: 3). 19. Fraser (1892: 164), notes on the ‘Samoan story of Creation’: ‘With equal kindness he lent us Rev. Mr. Powell’s original MS. text in the Samoan language, which is reproduced here after correction by the Rev. S. Ella of Sydney and himself. The Samoan text was not published by the Royal Society, but we have been induced to produce it here, for the same reason that other papers have been printed in the native languages in this Journal, i.e. in order that it may be read by the natives themselves. We thereby hope to induce members of the native races to contribute original matter bearing on their traditions, it appears to us that this “Story of Creation” is of a high order, and may be classed with the best of the creation myths of other branches of the Polynesian race. The Rev. T. Powell in securing this valuable tradition, the Rev. G. Pratt in translating it, and Dr. J. Fraser in editing and annotating it, have conferred a lasting benefit on the Samoans in particular and the Polynesians in general, which the descendants of the present people will as time goes on, appreciate more and more when education and refinement increase amongst them’. The word mamao is translated here as ‘limited extension’ and also space, Pratt (1893: 206) defines mamao as ‘v. to be far off, to be distant’. 20. Shore (1977: 438, see also 434) provides an account of the origin of the term tulāfale: the Tui Manu‘a (High Chief of Manu‘a) sent men to retrieve a fale that had been stolen by an atafu (raiding party). They returned with the fale propped on their shoulders. The Tui Manu‘a announced that, henceforth, they would be called tulaga o le fale (supports of the fale). 21. The ali‘i/tulāfale constellation is important to understand nineteenth-century Samoan politics (Tuagalu 1988: 7). Informal disussions between ali‘i and their representative tūlafale are more likely to lead to a consensus position, which is then presented to the alofisā of chiefs for further discussion and amendment, until a village consensus is reached.

50  I ‘uogafa Tuagalu



22. See Barrow (1968: 50). The Māori whare nui (meeting house) is the body of an ancestor – the rafters are his ribs and the ridge beam his spine – whereas the Samoan fale is modelled on the cosmos. 23. As a speculative aside, the mythical division of the earth and sky also resulted in spaces for domestic activities by the children of Earth and Sky, which is apt, given that the faletele and faleafolau also serve as the domestic centres for the chiefs’ domestic activities. 24. These changes in Samoan worldview, from pre-contact to the end of the German colonial period, are the subject of my PhD thesis (currently pending), tentatively titled ‘The Ontology of the Samoan Concept of vā, and the Evolution of the 19th Century Samoan Worldview’.

References ‘Action at a Distance’. 2015. In Jonathan Law and Richard Rennie (eds), A Dictionary of Physics, 7th edn. Oxford: Oxford University Press. Afoa, Ioane Asalele Lefauaitu. 2012. Taeao Mua: Le lomifefiloi o le Tusi Paia (Samoan Bible Dictionary). Nelson: The Copy Press. Aiono, Fanaafi Le Tagaloa 1997. O le Fa’asinomaga: Le Tagata ma Lona Fa’asinomaga. Alafua: Lamepa Press. Allen, Sophie. 2005. ‘Field Theory’, in Ted Honderich (ed.), The Oxford Companion to Philosophy, 2nd edn. Oxford: Oxford University Press. Anae, Melani. 2010. ‘Research for Better Pacific Schooling in New Zealand: Teu le Vā – a Samoan Perspective’, MAI Review 1. Barrow, Terence. 1968. The Decorative Arts of the New Zealand Maori. Wellington: A. H. & A. W. Reed. Bennardo, Giovanni. 2009. Language, Space and Social Relationships: A Foundational Cultural Model in Polynesia. Cambridge: Cambridge University Press. Bourdieu, Pierre. 1985. ‘The Social Space and Genesis of Groups’, Theory and Society 14: 723–44. Colman, Andrew. 2015. ‘Theory of Mind’, in Andrew Colman (ed.), A Dictionary of Psychology, 4th edn. Oxford: Oxford University Press. ‘Field’. 2015. In Jonathon Law and Richard Rennie (eds), A Dictionary of Physics, 7th edn. Oxford: Oxford University Press. Fraser, John. 1892. ‘A Samoan Story of Creation: A “Tala”’, The Journal of the Polynesian Society 1(3): 164–89. . 1897. ‘Folk Songs and Myths from Samoa II: Chaos and Strife: A Solo’, The Journal of the Polynesian Society 6(1): 19–36. Frazer, James George. 1967. The Golden Bough: A Study in Magic and Religion (Abridged edition). MacMillan & Co. Gilson, Richard Phillip. 1970. Samoa: 1830–1900: The Politics of a Multi-Cultural Community. Oxford: Oxford University Press.



The Energetics of Vā and the Samoan Faletele  51

Grey, George. 1965. Polynesian Mythology: And Ancient Traditional History of the Maori as Told by Their Priests and Chiefs, ed. Willam W. Bird. Christchurch: Whitcombe and Tombs. Gupta, Anil. 2021. ‘Definitions’, in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. Retrieved 26 November 2021 from https://plato.stanford.edu/ archives/sum2021/entries/definitions/. Hashimzade, Nigar, Gareth Myles and John Black (eds). 2017. A Dictionary of Economics, 5th edn. Oxford: Oxford University Press. Hesse, Mary B. 2005. Forces and Fields: The Concept of Action at a Distance in the History of Physics. Mineola: Dover Publications. Hilgers, Mathieu and Eric Mangez. 2015. ‘Introduction to Pierre Bourdieu’s Theory of Social Fields’, in Mathieu Hilgers and Eric Mangez (eds), Bourdieu’s Theory of Social Fields: Concepts and Applications. Milton Park: Routledge, pp. 1–40. Jammer, Max. 1993. Concepts of Space: The History of Theories of Space in Physics. Mineola: Dover Publications. Ka‘ili, Tēvita O. 2005. ‘Tauhi vā: Nurturing Tongan Sociospatial Ties in Maui and Beyond’, The Contemporary Pacific 17(1): 83–114. Kearney, Richard. 2003.  Routledge History of Philosophy Volume VIII: Twentieth Century Continental Philosophy. Milton Park: Routledge. Keesing, Felix M. and Marie M. Keesing. 1973. Elite Communication in Samoa: A Study in Leadership, 2nd edn. Redwood City: Stanford University Press. Krämer, Augustine. 1994. The Samoa Islands: An Outline of a Monograph with Particular Consideration of German Samoa: Vol. 1: Constitution, Pedigree, and Traditions (trans. Theodore Verhaaren). Auckland: Polynesian Press. Lafai-Sauoaiga, Solomon. 2000. Mavaega i le Tai. Apia: Methodist Printing Press. Le Mamea, Misi Kirifi, Te‘o Tuvale, T.E. Faletoese and F.L. Kirisome. 1958. O le Tusi Fa‘alupega o Samoa. Apia: Samoan Church Press (L.M.S.). Le Tagaloa, Fanaafi Aiono. 2003. Tapua‘i: Samoan Worship. Apia: Malua Printing Press. Lefebvre, Henri and Donald Nicholson-Smith. 1991. The Production of Space, Vol. 142. Oxford: Blackwell. Lehman, Frederic K. and David J. Herdrich. 2002. ‘On the Relevance of Point Field for Spatiality in Oceania’, in Giovanni Bennardo (ed.), Representing Space in Oceania: Culture in Language and Mind. Canberra: Australian National University Press, pp. 179–97. Levinson, Stephen C. 2003.  Space in Language and Cognition: Explorations in Cognitive Diversity. Cambridge: Cambridge University Press. Meleisea, Malama. 1987. The Making of Modern Samoa: Traditional Authority and Colonial Administration in the Modern History of Western Samoa. Suva: Institute of Pacific Studies of University of the South Pacific Press. Mila–Schaaf, Karlo and Maui Hudson. (n.d.). ‘Negotiating Space for Indigenous Theorising in Pacific Mental Health and Addictions’, Le Va.

52  I ‘uogafa Tuagalu



Milner, George. B. 2003. Samoan Dictionary. Samoan-English, English-Samoan. Auckland: Pasifika Press. Mosel, Ulrike and Even Hovdhaugen. 1992. Samoan Reference Grammar. Oslo: Scandinavian University Press. Nicholson, James. 2014. ‘Space’, in The Concise Oxford Dictionary of Mathematics, 5th edn. Oxford: Oxford University Press. Papaāli’i, Semisi Ma‘ia‘i. 2010. Tusi upu Samoa: Samoan to English, Vol. 1. Auckland: Little Island Press. Powell, Thomas. 1887. ‘A Samoan Tradition of Creation and the Deluge’, Journal of the Transactions of the Victoria Institute 20: 145–75. Pratt, George. 1862. A Samoan Dictionary: English and Samoan, and Samoan and English, with a Short Grammar of the Samoan Language. Apia: London Missionary Society’s Press. . 1893. Grammar and Dictionary of the Samoan Language SamoanEnglish, English-Samoan. Auckland: R. McMillan. Premack, David and Guy Woodruff. 1978. ‘Does the Chimpanzee have a Theory of Mind?’, The Behavioral and Brain Sciences 4: 515–26. Refiti, Albert. 2014. ‘Mavae and Tofiga: Spatial Exposition of the Samoan Cosmogony and Architecture’, PhD dissertation. Auckland: Auckland University of Technology. Shirali, Satish and Harikshan L. Vatsudeva. 2006. Metric Spaces. London: Springer-Verlag. Shore, Bradd. 1977. ‘A Samoan Theory of Action: Social Control and Social Order in a Polynesian Paradox’, PhD dissertation. Chicago: University of Chicago. . 1982. Sala‘ilua: A Samoan Mystery. New York: Columbia University Press. . 1989. ‘Mana and Tapu’, in Alan Howard and Robert Borofsky (eds), Developments in Polynesian Ethnology. Honolulu: University of Hawaii Press, pp. 157–73. . 1996. Culture in Mind: Cognition, Culture and the Problem of Meaning. Oxford: Oxford University Press. Statt, David A. 1998. ‘Psychology’, in The Concise Dictionary of Psychology, 3rd edn. New York: Routledge. Steubel, Oscar. 1896. Samoanische Texte. Berlin: D. Reimer. Tofaeono, Tavale Tanuvasa. 2012. Fa‘alupega o Samoa Atoa. Auckland: Fuelavelave Press. Tomlinson, Matt and Ty P. Kawika Tengan (eds). 2016. New Mana: Transformations of a Classic Concept in Pacific Languages and Cultures. Canberra: Australia National University Press. Tuagalu, I‘uogafa. 1988. ‘Mata‘afa Iosefo and the Idea of Kingship in Samoa’, Masters dissertation. Auckland: University of Auckland.



The Energetics of Vā and the Samoan Faletele  53

. 2008. ‘Heuristics of the Vā’, AlterNative, Special Edition (Special Issue): 108–26. . 2015. ‘A Typology of Space, the Samoan Concept of va and the Samoan Fale’, European Society for Oceanists, 10th Conference, 24 June 2015. Brussels, Belgium. . 2018. ‘The Energetics of the vā: Implications for the Hau of the Gift’, European Society for Oceanists: 2018 Confererence: Dealing with Inequality: Pacific Perspectives, Pacific Futures, 7 December 2018. London and Cambridge, United Kingdom. Tui Atua, Tupua Tamasese Ta’isi Tupola Tufuga Efi. 2009a. ‘Samoan Jurisprudence and the Sāmoan Land and Titles Court: The Perspective of a Litigant’, in T. Sua’alii-Sauni, I. Tuagalu, N. Kirifi-Alai and N. Fuamatu (eds), Su’esu’e Manogi: In Search of Fragrance: Tui Atua Tupua Tamasese Taisi and the Samoan Indigenous Reference. Apia: Centre for Samoan Studies, National University of Samoa, pp. 153–72. . 2009b. ‘Bioethics and the Samoan Indigenous Reference’, International Social Science Journal 60(195): 115–24. . 2015. ‘“Le Fuia, le Fuia, e Tagisia lou Vaelau: Starling, Starling, we Pine for your Nimbleness”: Towards a Samoan Indigenous Framing of Responsibility for Climate Change’, European Society for Oceanists, 10th Conference, 24 June 2015. Brussels, Belgium. Tusi Fa‘alupega Committee. 1985. O le tusi Fa’alupega o Samoa atoa. Apia: Methodist Church in Samoa. UNESCO. 1992. The Samoan Fale. Apia: UNESCO Office for the Pacific States and UNESCO Regional Office for Asia and the Pacific. Van der Ryn, Micah. 2005. ‘The Measina as Architecture in Samoa – An Examination of the va in Samoan Architecture and Socio-Cultural Implications of Architectural Changes’, The Measina a Samoa Conference III, 12–14 December, 2005. Apia: National University of Samoa. . 2016. ‘Which Way is Front?: Spatial Orientation Complications in Contemporary Samoan Villages’, Structure and Dynamics: EJournal of Anthropological and Related Sciences 9(1). Wendt, Albert. 1996. ‘Tatauing the Post-colonial Body’, Span 42/43: 15–29.

4

Vā and Its Relationship to the Samoan Built Environment ♦l♦

Anne E. Allen

Here [in Samoa] vā is so built into daily etiquette and protocols that it is like the air we breathe. Like, where you sit at a meeting, and how you arrange yourselves in the room at prayers or family meetings, who you speak to or not, etc. —Penelope Schoeffel Meleisea, personal communication1 Space is permeated with social relations; it is not only supported by social relations, but it is also producing and produced by social relations. —Henri Lefebvre, The Production of Space2

Introduction

A

mong the spatial concepts expressed in Sāmoa, the one most frequently emphasized in thought and action is vā. Vā may be defined, in part, as a physical distance. It can also denote being divided or separated and, thereby, establishing opposition of place with an intervening distance. This same word in Samoan also incorporates socio-spatial relevance in its general meaning as relationship. Each of these connotations comes into play in the conceptualization and utilization of vā in Samoan life. In her consideration of research into the art and aesthetics of Polynesian peoples, Adrienne Kaeppler asks, ‘How can we as outsiders understand Polynesian



Vā and Its Relationship to the Samoan Built Environment  55

worlds if we rigidly separate categories? Can we comprehend social and cultural patterns without understanding the layout of space, how one moves in it, what one wears while moving, and how all of these elements change according to contexts and activities?’ (1989: 211). However, I would add to this observation the need to consider vā as a dynamic mechanism, rather than just the physical frame in which other events and processes occur. This chapter draws from my work in 1990–1991, in the Itu-o-Tane district on the island of Savai‘i, Independent Sāmoa, as part of my art history PhD dissertation, as well as additional observations of Samoan life over the last thirty years. Most of my research has been concentrated in the village of Fagamalo and neighbouring communities. While my dissertation, entitled ‘Space as Social Construct: The Vernacular Architecture of Rural Samoa’ (1993), concerned itself with how space is conceived and expressed in village architecture, the concept of vā, although noted, was not a central focus. Since then, this concept, so vital to Moana cultures, has become the subject of intense investigation. In this chapter, therefore, I am directing a more narrow lens to consider whether the transdisciplinary application of Henri Lefebvre’s influential theories of space, and to a lesser extent Edward Casey’s ideas on place, can generate new insights into the Samoan built environment, its modes of production, its structure, and the actions that take place there. Thus, this inquiry is an attempt to bring Western thought to conceptual and physical Samoan expressions of space, to offer alternative insights into the dynamic processes that make vā such a vital component of the lived culture. As a common concept across many Pasifika cultures, vā is the arena in which relationships are formed and strengthened.3 This idea finds expression in the Samoan village and its built environment, as well as the ritualized actions that take place there. Within the structured architectural forms of a Samoan community, centralized ‘sacred space’ is established, and societal formation and reformation occur, a process that is intrinsically dynamic. The French philosopher Lefebvre, in his attempt to create a ‘unitary theory of space’, provides a potentially useful vocabulary and structure with which to explore vā.4

Theoretical Framework Within the larger Marxist dialectical process, examining any one aspect of society produces insight into other areas. Lefebvre brings together physical space (nature), mental space (formal abstractions about space) and social space (the realm of human interaction). In doing so, he suggests a heuristic triad that may be useful to a consideration of vā: spatial practice,

56  Anne E. Allen



representations of space, and representational spaces (Lefebvre 1991: 33). He ‘sought to unite three “realms” – the mental, social, and physical – into a conceptualization of what space is’ (Pierce and Martin 2015: 1282). This triad is not intended as a rigid structure or typology, but a dialectical simplification where physical, mental and social spaces blur into each other (Merrifield 2006: 109). Physical space coalesces with spatial processes in at least three basic ways: in its conceptualization, its actualization and its eventual transformation. For Lefebvre, these three exist in a dialectical relationship between the perceived, conceived and lived.5 Representations of space are ‘tied to the relations of production and to the “order” which those relations impose, and hence to knowledge, to signs, to codes and to “frontal” relations’ (Lefebvre 1991: 33). For our purposes, this is social production, the creation of relationships, not Marxist material production. It is conceptualized space implying abstract models found in concepts of ideal spatial organization – in Merrifield’s words, ‘what is in the head rather than in the body’ (2006: 109). Lefebvre considers this a conceived space, with ideology, power and knowledge embedded within its representation. Spatial practice embraces ‘social production and reproduction and the particular locations and spatial sets characteristic of each social formation’ (Lefebvre 1991: 33) and includes ideas and implementation. It is in this way that space is actualized, made real in a physical sense and, thus, perceived. ‘The spatial practice “secretes” that society’s space; it propounds and proposes it, in a dialectical interaction’ (Lefebvre 1991: 38). Lefebvre notes that such systems include routes, distinctive landmarks, paths and the boundaries that aid or deter a person’s sense of location and the manner in which they act. For Sāmoa, these elements are found in the layout of a village and that of family compounds, as well as in architectural types. Most importantly for a consideration of vā, Lefebvre includes the interactions that connect places and people and provides the setting for societal transformation as it is embedded in relationships. Representational spaces embody ‘complex symbolisms, sometimes coded, sometimes not’ (Lefebvre 1991: 33). It is space as directly lived through its associated images and symbols (Lefebvre 1991: 39). It overlays physical space, making symbolic use of its objects. ‘It embraces the loci of passion, of action and of lived situations and thus immediately implies time. Consequently, it may be qualified in various ways: it may be directional, situational or relational, because it is essentially qualitative, fluid and dynamic’ (Lefebvre 1991: 42). Kinkaid argues that, in his inclusion of space as lived, Lefebvre is providing us with a phenomenological interpretation that centres the body and its experience as the ‘key site in the production of space’ (2020: 174). But representational spaces are also



Vā and Its Relationship to the Samoan Built Environment  57

‘redolent with imaginary and symbolic elements, they have their source in history – in the history of a people and in the history of each individual belonging to that people’ (Lefebvre 1991: 41). Returning to Kinkaid, ‘The work such bodies unwittingly and intentionally do – marked and moved as they are by the contradictions and incomplete hegemony of ideological space – might prove powerful enough to transform space as we know it’ (2020: 184). Such a dynamic transformation incorporates both the element of time and the component of agency. I argue that this aspect can be observed in Sāmoa, when societal transformation occurs through the acknowledgement and creation of vā. Vā is fundamental to social existence. Individuals are embedded in space, in an environment in the broadest sense, built for and by the society in which they live. Although acknowledging the physical reality of a given space, Pacific peoples often give priority to what happens in it, and why. Studying this more dynamic aspect of vā expands beyond just the physical dimensions into a consideration of the actual processes of space found in the spatial practices of real communities. Lefebvre’s triad can be used to examine the conceptual mythos of space in various societies and, in turn, to provide insight into vā as a mechanism embedded also in the abstract models of Samoan architectural ideals.

Models of the Built Environment Representations of space occur in the conceptualization of physical spaces in Samoan thought. However, it is almost impossible to discuss such models without reference to Samoan spatial practices. Pierce and Martin note that places ‘are inherently, irreducibly hybrid. They are composed of disparate parts that are simultaneously discrete and combinatory; their borders and their internal segmentation are real but also provisional and unstable’ (2015: 1288). Further, Casey points out that, ‘[i]n creating built places, we transform not only the local landscape but ourselves’ (2009: 111). For this reason, any consideration of Samoan architecture and architectural layout is inherently also a discussion of vā, insofar as the built environment establishes the stage where vā takes place.

Family Compound Model Each ‘aiga (Samoan descent group) owns one or more areas of land within the village proper. Various structures in the family compound are arranged with a distinct front-to-back orientation and a subsequent public-vs-private differentiation. Radiating from the front, these are, respectively, guest fale,6 residences and then out buildings. In Samoan language and

58  Anne E. Allen



discourse, low status or subordination in general is associated with the ‘back’. Prestige is linked to the ‘front’. This distinction can be seen in both ceremonies and in the arrangement of architecture. The more prestigious structures are positioned at the anterior of the property. If a compound has a guest fale, it will be placed in the front, with the residential structures behind it. This is true whether the houses are open falesāmoa, enclosed falepālagi,7 or a mixture of both. If there is no guesthouse, residences will be built in the prestigious front area, often fully enclosed houses, fabricated by a Western-style contractor. Such falepālagi are linked to ideas of family wealth and access to outside resources. Behind sit the lowlier open residences, often faleo‘o or faleapa.8 To the back of the property are falekuka9 and other outbuildings that are fundamental to the working of the household, while at the same time being connected to heavy work or dirt. When I first moved to Savai‘i in 1990, it took some time before my adopted Samoan family became comfortable with my hanging out at the falekuka to chat or help with chores. Shore recounts a similar experience. Eventually, I did negotiate the nether regions of the village, and would make the same delicate transition in other villages in which I was to live. But always it took time. I came to realize that my village was a rather complex symbolic landscape. There were all sorts of invisible boundaries, differentiated social spaces that had to be taken into account if ever I were to make myself a home in Samoa. (1996: 268)10

The distinction between public and private spaces is a major factor in Samoan socio-political organization and corresponds with spatial practice and representational spaces within vā. The placement of the most prestigious structures in the most distinguished location reinforces each as a sign of family status, though the linear hierarchy is only imperfectly expressed in reality. Nevertheless, the general orientation remains as an important organizational strategy for the positioning of architecture within village spaces. Until the mid-20th century, the focus of public prestige in Samoan villages was a common open area called the malae, which was bordered by high-status guest fale. In the past, the main pathway into a community cut through this space, and visitors would thus first encounter the ‘front’ of each compound. Today, the public road may bypass the malae, providing another focus for prestige architecture (Allen 1993b: 37). Just as many in the United States will mow the publicly seen front lawn before the less visible back, Samoans are concerned with how the family is presented to passers-by. The linear pattern of front-to-back remains, although the focus may have shifted, and a strong ideal of centrality for the village endures, as expressed in the malae.



Vā and Its Relationship to the Samoan Built Environment  59

Figure 4.1. The linear hierarchies in fono seating as prestige vectors and nodes. © Anne E. Allen.

Village Spatial Models As the site for important religious and community activities, the malae, as both a symbolic and physical vā, provides the social and political core for the village. When in use for religious or political interactions, the malae becomes controlled space, where behaviour and entrance are restricted. Although material structures are generally absent on the malae, its physical existence is determined by the buildings that surround it and, therefore, mark its boundary. The malae, with its border of guest fale, is a spatial icon of the status relationships between families within the village. Consequently, the model of an ideal village is the malae as a hub, with a secondary ring of the guest fale surrounding the centre, and the smaller residences and outbuildings radiating farthest from the core (Allen 1993a: 271; Shore 1982: 50). However, the malae as literal spatial village centre is a rarely attained standard. Most Samoan communities outside the more ‘urban’ areas around the capital Apia or the villages surrounding the wharf at Salelologa are more linear in form, extending along the coastline and divided by a road. The narrow area between beach and mountains prevents the physical realization of the circular ideal. For inland villages, there may be restraints by mountains, rivers or other geographical features. In some villages, the existence of multiple malae results in several foci, manifesting the political realities of those communities.11 Even in the urban sprawl that is modern Apia, individual suburban villages retain a malae in some form. Although in most instances the centralized ideal is imperfectly instantiated, the concept of an open area for public events is retained.12

60  Anne E. Allen



The physical bracketing that defines the vā of most villages is made up of those boundaries (either visual or implied) with the sea, the bush or other villages. Shore therefore postulates a second, linear conceptual model for Samoan community space in which villages, too, are conceived of as having front (luma) and back (tua) areas. The front of the village is the side facing the sea (tai), while the back is the bush side (uta) (Shore 1996: 269).13 Tai suggests the ocean, the open coast and the vā of intense social contact. In the past, visitors travelling any significant distance typically arrived from the sea. In the Samoan creation myth from Manua, the sea is the original space that allows for the creation of vā.14 The sea as a connecting vā that facilitates relationships is also at the core of Epeli Hau‘ofa’s 1994 essay, ‘Our Sea of Islands’. For Samoans, tai conceptually relates to organized human life, proper behaviour and the control of the village chiefs. In contrast, uta implies a-social conduct, since human activities inland are unimpeded by the rules set forth by the chiefs (Shore 1996: 269). According to Shore, the sea vs. inland opposition, where geographic features are used for ‘mapping social, kinesthetic, and moral attributes, appears to be a fundamental orientational framework for Polynesians. … The “back” for Samoans, as for Polynesians more generally, is associated with low rank, and with impulsive rather than socially correct behavior. The “front”, by contrast, implies high rank, social authority, and socially visible and hence constrained behavior’ (Shore 1996: 269), making tai and uta moral orientations. Such a linear model might seem incompatible with the centralized ideal in spatial practice, but the concentric prototype does incorporate equivalent ‘outer’ areas, either bush or sea, in terms of their relation to the centres of human political life (Shore 1996: 272). Shore calls the linear concept an ‘explicit cultural model’, the concentric model a ‘tacit cultural model’ (1996: 272– 73). However, the construction of prestigious architecture facing the public road suggests there is a third model at play in Samoan representations of space and the resulting spatial practices, beyond the tai-uta linear dualism postulated by Shore or the centralized model of vā.15 The correlation between private and public is based on fundamental Samoan concepts of social connections, or vā in its sense as both physical space and relationship. In Samoan speech, one may refer to o le pola motu i tua, the torn blinds at the back of the house. ‘Passers-by see only the front of the house; therefore, it is not considered necessary to pay any particular attention to the things at the back’ (Schultz 1985: 89). With the post-Independence development of extra-malae vehicular roadways came a change in the way in which visitors first experience a village. Instead of coming into the traditional community centre, in many cases they enter what at



Vā and Its Relationship to the Samoan Built Environment  61

one time would have been the less prestigious, privately orientated back. In this situation, the important discursive statements that are directed towards a dialogue with the larger society and embodied in the guest fale would be diluted or missed entirely by travellers. An architectural shift in spatial practice from the malae, as a single expression of vā, to the road, by families with land adjacent to the thoroughfare, ensures that visitors first encounter the more distinguished and appropriate buildings. The application of community rules to the road precinct (as well as the malae) reinforces the use of this space as a venue for public statements and interaction, a village vā. In contemporary communities, the structural distinction between front and back is retained but manipulated in regard to public orientation, rather than geographic location. The road establishes a focal line within the village in addition to the malae centre-point. The hierarchy of public over private is an additional aspect of representations of space, one that overlies, and in some ways supersedes, both the linear and centric models in actual spatial practices.

Vā as Representational Spaces Up to now, I have considered conceptual ideals that are part of Samoan representations of space and how these have affected spatial practices. All three models reflect established vā: relationships within the family and community, particularly relating to status and social control. Geertz argues ‘that human thought is basically both social and public – that its natural habitat is the house yard, the marketplace, and the town square. Thinking consists not of “happenings in the head”’ but of an interchange in meaningful symbols (Geertz 1973: 45). Socio-political relationships are not just signified, but actively maintained and transformed in the actual usage of built structures and the spaces they help define – in what Lefebvre calls representational spaces. This manifestation of vā as dynamic, rather than just as a static framing, occurs through the manipulation of place within architecturally defined forms, analogous to what Casey has termed the ‘generative power’ of place in its inclusion of space and time (1996: 43). Consequently, we move beyond the physical mapping of vā to add the component of active agency. It is nevertheless important to start with the map itself, the representation of space, as Lefebvre tells us that representational spaces ‘tend toward more or less coherent systems of non-verbal symbols and signs’ (1991: 39). And it is in symbolic ceremonial actions that relationships are expressed or formed within the architectural environment, and vā becomes both form and function.

62  Anne E. Allen



Vā as Ceremonial Space Within the cycle of Samoan village life, there are several common formal events. Such ceremonies share the establishment or reinforcement of social ties through the structuring and manipulation of vā within the architectural ‘landscape’. These rituals include a placement within, or the crossing of, an open centralized space, the vā. For example, the yearly inspection of mats, sheets and other household goods by the Women’s Komiti focuses on the open centre. Each item is displayed in turn and piled in the middle of the house. Although there is friendly competition between the participants, the inspection reinforces the organizational ties that the Komiti provides for the village women. When the community is gathering foodstuffs in preparation for a large ceremonial presentation, each family’s contribution is placed in the centre of the fale, called out and duly marked by those present. Traditional Samoan architecture has a single space and no interior or exterior walls. In everyday activities, landmarks, paths and boundaries are understood and acknowledged as people move about family compounds and the community. This is one aspect of the vā of spatial practice. However, routine actions are set aside in the context of rituals of social relationship, which generally occur on the malae or in individual guest fale, providing more overt expression of spatial practice. In Sāmoa, a fono is the political meeting of matai (chiefs in a Samoan village) (Duranti 1981: 1).16 Although each family is independent in its internal affairs, and thus each matai equivalent to other heads of families, there are differences of type and rank: ali‘i and tulāfale. The ali‘i are the so-called high chiefs. The tulāfale are the talking or orator chiefs. The distinction between the two chiefly types is reflected in Samoan representations of space as a conceptual model, but also expressed in language, action, place and space.17 Fono typically take place in the guest fale. However, when the meeting is one of profound sacredness or great importance, the village may come together on the vā of the malae itself. The seating arrangement in fono follows a basic pattern, with slight contextual variations. For all ceremonial meetings there must be a correspondence between the members of the group and their position … the social relationship is bodied forth in a seating plan with a definite symbolic value, and the seating plan is always conceived as a circle in which each segment has a definite value, and each is essential to the complete ceremony. (Mead 1930: 55)

In Lefebvre’s terminology, this plan is an example of spatial practice, articulating vā in its aspect as representational space and active mechanism. Within the Samoan guest fale, there are two pairs of focal points. One set consists of the centre posts at the ends, right and left relative to the



Vā and Its Relationship to the Samoan Built Environment  63

structure’s front. The other pair is comprised of the centre points of the anterior and posterior of the house. The ends are reserved for the two highest-ranking ali‘i.18 The lesser ali‘i (where applicable) seat themselves to either side of these. This flanking of the central figures with those of decreasing rank provides a linear arrangement. While a distinction is drawn between places to the left or right, the interior vā is also divided front to back. Such a division corresponds to that found within the family compound; the front is equated with public prestige while the back plays a support function. During a fono, the anterior of the house is reserved for the tulāfale. They arrange themselves in the front (like the ali‘i at the ends) with the most prestigious individuals in the centre. Those charged with the making and distribution of ‘ava, a support function, occupy the posterior of the fale. Here the taupou (ceremonial maiden) or the manaia (titled chief ’s son) is placed in the centre, with their helpers flanking and slightly behind them.19 An integration of the two (oppositions and linear hierarchy) results in a central open space, a vā, with defined borders. Space is thus established along linear hierarchies that employ these four focal points as apogees of prestige. As one moves away from these peaks in either direction, the status of locations lessens. At the same time, the house ends are more prominent than the centre points of either the front or back. The resulting set of relationships can be diagrammed as vectors flowing from positions of high to low prestige. (Figure 4.1, above) Although not marked by posts or constrained by the structure of the guest fale, such an arrangement is also expressed when a major event takes place on the malae. While distinct nodes or places are marked as prestigious, front, back, left and right have varying degrees of expression, resulting in a fluid model that is adaptable to specific contexts. For example, if the fale is crowded, the senior orators will take their places closer to the centre vā. Those of lesser rank will sit behind. All the locations in the front row are more prestigious than any in the back. Even at the ends, lesser ali‘i will position themselves at the back of their higher-ranking counterparts if there is not enough room to the sides. Even the ranking individual assigned with preparing the ‘ava is seated slightly to the front of the untitled individuals who assist. Since it is immaterial whether the order is along a straight line or a curve, this spatial practice is responsive to the shape of the individual house or malae. A border is created, resulting in the concentric opposition between the centre and periphery. In all circumstances, the more prestigious positions are closer to the physical core of the open interior space and away from the edges. A diagram of the positional interrelationships within the house can thus be configured like a topographical map (Figure 4.2). Although the landscape as depicted calls to mind the ridge of a crater with a collapsed centre, in

64  Anne E. Allen



reality it is the inverse. The middle is the highest point, resulting in something akin to a mountain peak. Such an ideogram of a formal meeting, with a central prestige apogee, corresponds to the hierarchical organization of the village model as a whole, with its centralized malae, no matter how inexactly expressed in the actual constructed space. The conceptualization of a circle as the periphery of a formal meeting suggests the equivalent status of the individual matai in their role as the heads of autonomous families. Yet, the circular form is overlaid by hierarchical considerations. Individual fono may show variations based on particular contexts; however, the overall pattern of periphery/centre and oppositive dualism20 remains an organizing spatial model.21 Such a seating plan corresponds with Lefebvre’s representations of space and establishes a distinction between centre and periphery. However, these larger sections of vā are in turn marked internally through the designation of places.

Place and Vā as Social Mechanism As we have just seen, Samoan representational spaces, social status, and thus the structuring of socio-political relationships, are communicated by place. Therefore, we need to consider place beyond its meaning as mere location, for social action subsequently involves breaking place and, in

Figure 4.2. Topographical ideogram of prestige in the vā established by seating positions. © Anne E. Allen.



Vā and Its Relationship to the Samoan Built Environment  65

effect, collapsing vā. In considering the concept of place, John Agnew summarizes the two stances typically taken in regard to the relationship of space and place. These two belong to an ‘abstract spatial analysis which tends to view places as nodes in space simply reflective of the spatial imprint of universal physical, social or economic processes’, on the one hand, and ‘concrete environmental analysis which conceives of places as milieux that exercise a mediating role on physical, social and economic processes and thus affect how such processes operate’ (2011: 2). From the first perspective, place is conceived merely as a geometric part of space; from the second, place is understood phenomenologically as a ‘distinctive coming together in space’ (2011: 2). ‘Place as node’ is evident in the seating arrangement in fono and in the spatial practices shaping the architecture of family compounds and village organization. However, in Casey’s view (2009), it is vital that we consider the actions of staying in place, moving within and between places, as well as the particulars of context. Thus, place-making (or marking) is ‘the set of social, political and material processes by which people iteratively create and recreate the experienced geographies in which they live … Place-making is an inherently networked process, constituted by the socio-spatial relationships that link individuals together through a common place-frame’ (Pierce, Martin and Murphy 2011: 54). In Sāmoa, place acts dynamically within a vā, rather than just belonging to a framework in which events and processes take place. In the rituals found in Samoan life, coming together across the vā created by architecture and the built environment, either literally or symbolically, acts on social relationships. The house is already inscribed with the body politic of the village, so that whoever is present has already been identified and marked in the house with their vā relationships by virtue of who they are and what or whom they (re)present. In this sense the house is almost an anticipating vā machine, a stage in need of actors that when not in use lies dormant waiting, anticipating bodies that are charged with vā relationships which may activate the openness that fold[s] bodies together in a boundless vortex that suck[s] together the in-between relations of people. (Refiti 2008: 100)

Although the following discussion will focus on the house as locus of events, this potential of spatial vā to expedite the social vā is also found in the open, village malae.

The ‘Ava Ceremony The signification of status and rank hierarchy in the spatial placement of any formal assembly, in either the guest fale or the malae, is reinforced

66  Anne E. Allen



through the actions taken in the ‘ava ceremony, which occurs at most important occasions. The ritual includes everyone seated in a circular pattern and concerns the creation or expression of social ties. Those who receive kava are called out in order of status by one of the tulāfale. Once a name is identified, the cupbearer raises the ‘ava above his head, steps into the central vā of the house and pauses before turning and advancing to the person being served. When he arrives at the recipient, with a graceful sweeping movement the bearer hands the cup. The first person to receive the drink is the highest chief of any visiting party, who is followed by the ranking matai of the host party or village. The leading tulāfale of the visiting party is then served the ‘ava, followed by the leading orator of the village and so on, all in order of rank. Doreen Massey’s general observations on place also apply to the Samoan ‘ava ceremony. ‘Places may be thought of as open articulations of connections’, and ‘identities of subjects and identities of places constructed through interrelations not only challenge notions of past authenticities but also hold open the possibility of change in the future’ (1999: 288). Both the possibility and actuality of change are fundamental aspects of vā as a dynamic of social relationship formation. In fono and ‘ava ceremonies, place, determined by rank and position, reflects those aspects of Samoan socio-political relationships. However, it is in the crossing of the vā during the presentation of the cup that the participants are woven together in a concrete semiotic fashion. Social relationships are made manifest, as well as forged, in formal prestations at public ceremonies.

Presentation of Gifts In Sāmoa, social connectedness defines one’s identity as much as individual characteristics – if not more so (Shore 1982: 136–37). However, relationships are not fixed equations, but are modified through actions. The conferral of textiles and other gifts in architecturally defined spaces is a major component of Samoan ritual activity. Activities, such as giving gifts, the sharing of food and dance performances, can produce and strengthen relationships.22 Formal transferal of property occurs at all fa‘alavelave (life crises ceremonies) and most public occasions. Fa‘alavelave is used to indicate an accident, fight, or any trouble, as well as social ceremonies. Pratt defines the word as a hindrance or impediment, while the root lavelave indicates entanglement (1977: 93) and stresses the social relationships that underlie Samoan ceremony. Although the specifics of each event are unique, the actions of bestowal do conform to a general pattern. The gifts are carried in public display, across



Vā and Its Relationship to the Samoan Built Environment  67

the open vā of the house or the malae. The goods themselves form part of a reciprocal system that binds distinct social groups (Allen 2007a). Keesing and Keesing note that ‘the rights and obligations generated through property exchanges could be used to cement alliances or to redress a balance in case of feuding or war… Even under modern conditions of peace, they are likely to be called upon to repair breaches in social relationships’ (1956: 77). The goods themselves form part of a reciprocal system that binds distinct social groups (Allen 2007a) and symbolize the relationships involved. However, it is the passage through vā, I would argue, that transforms them from mere matter to carriers of semiotic significance. Thus, vā moves from being the shell for such social formation to an active process. Whether on the malae or within the guest fale, the groups involved in prestations sit across from one another, separated by the central vā. The orator chief representing the family or village loudly proclaims each gift. As the goods are brought forth and carried across to the recipients, the transfer creates a conceptual link between the spatial locations of the participants. By crossing the vā, the gifts and their movements link the participants’ physical places and act as agents of socio-political communication (Allen 2007a). This modification of spatial relationships is emphasized when the gifts are lengths of tapa or modern cotton cloth. The textiles are displayed fully open and, in their movement, provide a visual connection between the participating groups. When full bolts of cloth are presented, they often form a physical link concretizing the conceptual merging of places within the vā, as well as the relationships that are created or reinforced. The distributions at fa‘alavelave function as both social and economic transactions, actively binding individuals and groups together within a system of reciprocity. ‘Display and transfer of property in gift-giving ceremonials may become a highly developed form of communication among groups and their elites and an important technique for demonstrating and validating status and for fostering social integration’ (Keesing and Keesing 1956: 76). Within Samoan formal presentations, the messages are statements concerning social space and its ordering. The gifts become ties, which cross architecturally defined space and nullify the vā of spatial practice between families. Both actions and objects reflect the already existing structure and modify it through the revising of relationships and their relative positions within the larger society. This is accomplished through the visual presence and physical attributes of the transferred articles, as well as their movement through vā as defined by physical architecture, which results in a dynamic re-organization of social reality.

68  Anne E. Allen



Conclusion In Sāmoa, vā and its expression in the built environment of the village is fundamental to the establishment and maintenance of social relationships. Vā is not a space, like a void, where things can happen, but a dynamic agent in the organizing of community and individual relations. Vā is both a unitary and unifying concept and an active mechanism of social relationships. Kaeppler has suggested that the underlying organizational axioms within a culture can be ‘recognized by members of the society as a set of principles with which they helped to organize their lives – not necessarily verbalized as such but derived from ethnographic data in several domains in which the structure consistently repeats itself ’ (1978: 261). Vā is repeatedly manifested in architectural forms and in the everyday and ritualized actions that take place in these settings. It finds expression in the spatial organization of several facets of Samoan life: conceptual, physical and social. Alternatively, in the terminology of Lefebvre, vā is a fundamental component of representations of space, spatial practice and representational space. Shore has argued for two models of spatial practice regarding the vā of the Samoan village, one centralized, the other linear. I suggest a third, the dualism of public vs. private. The first two are ideals that are seldom fully recognized in the built environment, due to competing typographies and the exigencies of local histories. By contrast, the distinction between public and private is consistently found in actual praxis and in the daily application of the philosophical concept of vā. All three models are found in formal events, such as fono or the ‘ava ceremony, with their emphasis on linear hierarchies, and the polarity of placement in regard to front vs. back, which is essentially a spatial dualism of public vs. private. The seating arrangement at a chiefly council within the Samoan built environment thus functions as an ideogram of the social relationships within the larger village vā. As a representation of an acknowledged hierarchy, the fono is organized around focal points at the periphery and reinforces this hierarchy through the activities of the server within the ‘ava ceremony. Although incorporating movement, these activities emphasize the established places as nodes. However, Samoan social transformation occurs via representational space and the manipulation of place within the vā defined by this relatively static placement. When paired with human actions, both vā and the places within it become active agents in the evolution of relationships within the Samoan socio-political sphere. Such ties are established or maintained between distinct families through the everyday movement of goods and food across the known, if not marked, boundaries between land and architecture. The



Vā and Its Relationship to the Samoan Built Environment  69

formal presentation of gifts ritualizes these transfers by reconfiguring vā via the conflating of place across the open centre. Lefebvre asks, ‘[w]hat term should be used to describe the division which keeps the various types of space away from each other, so that physical space, mental space, and social space do not overlap? … In actuality each of these … kinds of space involves, underpins and presupposes the other’ (1991: 14). Without the representations of space, which are the product of the mind, spatial practices do not exist as a coherent expression. In turn, neither can result in the evolution of social relationships except as representational spaces. And, in Sāmoa, each of these united together is vā: stage, actor and social creation. Anne E. Allen is Professor of Fine Arts and has taught at Indiana University Southeast since 1994. She is a recipient of the Fulbright-Hays Scholarship, which funded a full year of research in the independent Pacific nation of Sāmoa. She returns there whenever possible to continue her work. Her research areas include Samoan village architecture, cloth and ritual, and how these are reflected in and facilitate cultural conceptualizations and usage of the spatial domain. She has contributed numerous publications to the field of Pacific Art History as well as international and national conference presentations. Working with students, Anne has curated several exhibitions for the Indiana University Southeast Ronald Barr Gallery of art from Africa, the Pacific and Native America.

Notes   1. Dr Penelope Schoeffel Meleisea is an anthropologist and Associate Professor at the Centre for Samoan Studies, National University of Samoa. She is married to the noted Samoan historian Meleisea Leasiolagi Dr Malama Meleisea.   2. Henri Lefebvre (1901–1991) was an important Marxist philosopher and social theorist. He wrote what are considered seminal works on the reprioritization of space in social and critical analysis.   3. For examples, see Amituanai-Toloa (2007), Anae (2010) and Ka‘ili (2005).   4. In utilizing Lefebvre, I am concerned with his conceptualizations and categorizations as a means of understanding vā, not with his focus on economics, neo-capitalism and urban spaces, in part because these subject areas are enmeshed in a Western capitalistic view that is at odds with traditional Samoan values and understanding of the world. I also reject the Marxist categorization of the primitive that is found in some of the examples Lefebvre uses. Thus, it is the use of Lefebvre’s vocabulary and basic concepts that facilitates analysis.   5. For a discussion of the influence of Lefebvre on the fields of philosophy and geography, see Pierce and Martin (2015).

70  Anne E. Allen



  6. The formal term for a guest house is faletalimalo. Faletalimalo are now made up of three types, each without exterior walls or internal dividers. Traditional faletele are oval with central posts supporting cross beams. Faleafolau are longer in proportion with sides posts bracing the roof joists. Today, there are also rectangular faleapa, which are created solely with Western carpentry and materials such as sawed lumber and tin roofs. Samoans in the Itu-o-Tane district generally call a faletalimalo a faletele in everyday discourse. However, since faletele also applies to a specific style of traditional architecture, to avoid confusion I will use the English/Samoan hybrid term ‘guest fale’.  7. Commonly, in my experience people differentiate enclosed houses as falepālagi, in contrast to the open faleapa, although both utilize non-traditional shapes, materials and technologies. In modern usage, pālagi denotes outsiders, particularly those from Western countries.  8. A faleo‘o is a small open structure, used for sleeping, typically built using local materials with either a tin or thatched roof. Stylistically, faleo‘o are often fale fa‘aivi‘ivi, that is lacking the structural crossbeams that are intrinsic to larger buildings created using traditional techniques and materials.  9. A falekuka is a separate, open building used for cooking. 10. The tension between honouring guests and allowing them to participate in everyday chores becomes evident whenever I bring students to Sāmoa. It is important to me that they do not take advantage of our Samoan hosts, a perhaps very Western concept, while at the same time acknowledging the Samoan tradition of hospitality. 11. Gratten (1948: 53) contends that there are actually two village forms: one with a central public meeting area and a second exhibiting a single aligned tract of houses facing the beach with the malae in front. However, I would suggest that the latter is actually a modification of the former, reflecting local topography and history. Most Samoan villages are coastal, although archaeological data suggests that this was not always the case. Where the mountains are close to the sea, the possibility of a completely central village space is eliminated by the narrowness of the coastal strand. Villages such as Manase on Savai‘i or Luatuanu‘u on ‘Upolu consequently exhibit a linear shape. Where more room exists, such as in Solosolo on ‘Upolu, the villages are often much more centralized in their configuration. 12. Since my original research in 1990–1991, the visual merging of communities in Apia and Salelologa, although not necessarily the physical or social integration, has intensified. In addition, more and more families are choosing to build away from the village core, even in distinctly rural areas. However, the malae as both conceptual and social centre is retained. 13. Shore’s earlier view that tai-uta is a physical distinction in terms of actual house placement based on Salailua is countered by Allen (1993a: 268–71).



Vā and Its Relationship to the Samoan Built Environment  71

14. Tagaloa the Messenger took on the form of the Tuli bird and went out in search of land, but only the waters could be seen. The Manu‘a group was induced to rise up, creating a new space in the now divided seas (Powell 1887: 156). 15. See also Van der Ryn (2012: 127–29). 16. The term fono can designate a variety of meetings. These include women’s committee meetings, those of the untitled men to discuss village work projects, and so on (Duranti 1981: 13 and 17). In informal speech, fono may designate a departmental meeting in business or government as well. I am using the formal arrangement seen in chiefly village meetings, whether the monthly scheduled ones or those called ad hoc to consider important community affairs that cannot wait. 17. For a more detailed explanation of the matai system, see Shore (1982) and Meleisea (1987). 18. Shore (1982: 79) specifies ali‘i (high chiefs) for these positions. In contrast, Mead (1930: 56) states that the right-hand site is reserved for the ali‘i while the left is occupied by the senior tulāfale. These differences probably reflect the hierarchies unique to each village. In Fagamalo, where there are no ali‘i, both spaces are occupied by talking chiefs. Duranti (1981: 57–60) gives an example where the high chief sat more to the side than the middle of the house end. As rank is related to location, the ali‘i was overtly lowering his position more in par with his fellow chiefs. In addition, the chief who sits in any particular space may vary with circumstances, the presence of different matai influencing the seating arrangement. As the scale of meetings expands from family to village to district level and beyond, the room for variation increases. ‘This general scheme for the fono is flexible enough to allow for a collapsing of distinctions, with one unit representing a general group… Conversely, the scheme allows for elaboration of fine distinctions’ (Shore 1982: 81). 19. Although the preparer is usually someone holding a taupou or manaia chiefly title, within the hierarchy of the ceremony they function to serve the assembly and are thus positioned at the back. Typically, their helpers are untitled men and are consequently farther away, if only slightly, from the centre vā. 20. Here I am using oppositive in both its meaning as expressing antithesis or opposition and in literal placement of corresponding positions with relation to an intervening space. 21. I observed a variation of this pattern at the ceremony honouring the installation of the Fagamalo Congregational Christian Church pastor as the district senior. In this case, the spatial arrangement was based on district hierarchies and accommodated four groups. The faife‘au (pastors) were seated to the left, with the senior (oldest) positioned at the centre. He was flanked by the other ministers in general order of their ages (and thus, time of service and rank). Across from this location sat the high chiefs of those villages with attending

72  Anne E. Allen



pastors. The Fagamalo matai, who are all tulāfale, were at the front of the house, facing the ‘ava preparer and her assistants. 22. Although not a focus of this chapter, for more on dance as a means of social formation, see Allen (2007b).

References Agnew, John. 2011. ‘Space and Place’, in John Agnew and David N. Livingstone (eds), Sage Handbook of Geographical Knowledge. London: Sage Publications, pp. 316–30. Allen, Anne. 1993a. Space as Social Construct: The Vernacular Architecture of Rural Samoa. PhD Dissertation, New York: Columbia University. . 1993b. ‘Architecture as Social Expression in Western Samoa: Axioms and Models’, Traditional Dwellings and Settlements Review 5(1): 33–45. . 2007a. ‘The Tie that Binds: Siapo, Western Cloth, and Samoan Social Space’, in Heather E. Young-Leslie and Ping-Ann Addo (eds), Hybrid Textiles: Pragmatic Creativity and Authentic Innovations in Pacific Cloth, Special Issue, Pacific Arts. NS, 3–5: 94–103. . 2007b. ‘Transforming Samoan Space into Place: Dance and the Construction of Community’, The International Journal of the Arts in Society 1(5): 1–10. Amituanai-Toloa, Meaola. 2007. ‘The “Va Tapuia” (Space Made Sacred) in Bridging Research and Relationships: Brown Culture and Commonsensical Ethics’, AlterNative: An International Journal of Indigenous Scholarship 1: 204–27. Anae, Melani. 2010. ‘Teu le Va: Toward a “Native” Anthropology’, Pacific Studies 33(2/3): 222–40. Brockelman, Thomas. 2003. ‘Lost in Place? On Edward Casey’s Anti-Modernism’, Humanitas 16(1): 36–55. Casey, Edward. 2009. Getting Back into Place. Second Edition. Bloomington: Indiana University Press. . 1996. ‘How to Get from Space to Place in a Fairly Short Stretch of Time: Phenomenological Prolegomena’, in Steven Feld and Keith Basso (eds), Senses of Place. Santa Fe: School of American Research Press; Distributed by the University of Washington Press, pp. 13–52. Duranti, Alessandro. 1981. ‘The Samoan Fono: A Sociolinguistic Study’, Pacific Linguistics. Series B (8) Australian National University Department of Linguistics, Research School of Pacific Studies, Australian National University. Epeli, Hau‘ofa. 1994. ‘Our Sea of Islands’, The Contemporary Pacific 6(1): 148–61. Geertz, Clifford. 1973. Interpretation of Cultures. New York: Basic Books. Gratten, Frederick. 1948. An Introduction to Samoan Custom. Apia, Samoa: Samoa Printing and Publishing.



Vā and Its Relationship to the Samoan Built Environment  73

Kaeppler, Adrienne. 1978. ‘Melody, Drone, and Decoration: Underlying Structure and Surface Manifestations in Tongan Art and Society’, in Michael Greenhalgh and Vincent Megaw (eds), Art in Society. London: Duckworth. . 1989. ‘Art and Aesthetics’, in Alan Howard and Robert Borofosky (eds), Developments in Polynesia Ethnology. Honolulu: University of Hawaii Press, pp. 211–40. Ka‘ili, Tēvita. 2005. ‘Tauhi Va: Nurturing Tongan Sociospatial Ties in Maui and Beyond’, The Contemporary Pacific 17(1): 83–114. Keesing, Felix and Marie Keesing. 1956. Elite Communication in Samoa: A Study of Leadership. Stanford, CA: Stanford University Press. Kinkaid, Eden. 2020. ‘Re-encountering Lefebvre: Toward a Critical Phenomenology of Social Space’, Society and Space 38(1): 167–86. Lefebvre, Henri. 1991. The Production of Space. Malden: Blackwell. Original work published in 1974. Mageo, Jeannette. 1989. ‘“Ferocious Is the Centipede”: A Study of the Significance of Eating and Speaking in Samoa’, Ethos 17(4): 387–427. . 2002. ‘Myth, Cultural Identity, and Ethnopolitics: Samoa and the Tongan “Empire”’, Journal of Anthropological Research 58(4): 493–520. Massey, Doreen. 1999. ‘Spaces of Politics’, in Doreen Massey, John Allen and Philip Sarre (eds), Human Geography Today. Malden: Polity Press, pp. 279–94. Mead, Margaret. 1930. The Social Organization of Manua. Honolulu: Bishop Museum Press. Meleisea, Malama. 1987. The Making of Modern Samoa: Traditional Authority and Colonial Administration in the History of Western Samoa. Suva: Institute of Pacific Studies of the University of the South Pacific. Merrifield, Andy. 2006. Henri Lefebvre: A Critical Introduction. Abingdon, UK: Routledge Publishing. Pierce, Joseph and Deborah G. Martin. 2015. ‘Placing Lefebvre’, Antipode 47(5): 1279–99. Pierce, Joseph, Deborah Martin and James T. Murphy. 2011. ‘Relational PlaceMaking: The Networked Politics of Place’, Transactions of the British Institute of Geographers 36(1): 54–70. Powell, Thomas. 1887. ‘A Samoan Tradition of Creation and the Deluge’, Journal of the Transactions of the Victoria Institute or Philosophical Society of Great Britain 20: 145–75. Pratt, George. 1977. Pratt’s Grammar and Dictionary of the Samoan Language. Apia: Malua Printing Press. Reprint of the 1911 Fourth Edition. Refiti, Albert. 2008. ‘The Forked Centre: Duality and Privacy in Polynesian Spaces and Architecture’, AlterNative: An International Journal of Indigenous Peoples 4: 97–106. Schultz, Erich. 1985. Samoan Proverbial Expressions: Alaga‘upu Fa‘a-Samoa. Suva: Polynesian Press.

74  Anne E. Allen



Shore, Bradd. 1982. Sala‘ilua: A Samoan Mystery. New York: Columbia University Press. . 1996. Culture in Mind: Cognition, Culture, and the Problem of Meaning. Oxford: Oxford University Press. Van der Ryn, Fepulea‘i Micah Gabriel. 2012. ‘“The Difference Walls Make”: Cultural Dynamics and Implications of Change in Samoan Architectural Traditions and Socio-Spatial Practices (1940–2006)’. PhD Dissertation, Auckland: University of Auckland.

5

‘Carving Costs Nothing’ Māori Woodcarvers Train Wage-Labourers How to Show Up to Work on Time ♦l♦

Jacob Culbertson

I would venture there is no identity of a practice independent of its environment. This emphatically does not mean that the identity of a practice may be derived from its environment … Spinoza might say to us, we do not know what a practice is able to become; what we know instead is that the very way we define, or address, a practice is part of the surroundings which produces its ethos. —Isabelle Stengers, ‘Introductory Notes on an Ecology of Practices’

The House of Knots

T

he centre of the universe is a Māori woodcarving shop in an old shoe factory in Aotearoa New Zealand’s Bay of Plenty, called the House of Knots. Four carvers work there daily. Each arrived at different points in the late 1990s and at different moments in their working lives, coming to the shop to enrol in a course in Māori carving. When I met them, the oldest was around sixty years old, a former Telecom lineman who retired after he busted his hip in a pig hunting accident. The youngest was thirty, with a degree in Māori arts from a big-city ‘polytechnic’ tertiary institution. Another was a minister in a prophetic church and an anime fanboy. The

76  Jacob Culbertson



fourth carver I can only describe as a washed-up street fighter, socially aggressive with a goofy charm, who found his peace among the tikanga (traditional protocols) of the House of Knots. For a time, this group called themselves an ‘artists’ collective’. The bulk of their work, by quantity, is comprised of carved paddles, masks, treasure boxes and the wooden keys that parents customarily give their kids on their twenty-first birthdays – all small things, affordable, easy to hang on the walls of a house and easy to bring to a party. A few times each year, a larger project comes through the shop. In my time, for example, I worked on a door lintel for a local history museum; a refurbished gateway for a meeting ground, for which we cut out the rot and added new paint; and at least ten poupou, the panels of ancestors rendered as serpentine manaia or humanoid tiki that line the walls of meeting houses. I first arrived at the House of Knots in 2006, introduced by an old friend who had completed the government-sponsored certification course in carving – a handy stopgap to get on his feet when he moved back to his ancestral lands after growing up in the city. At that time, the carving shop hosted a welfare-to-work scheme, run by the local tribal trust board with funds and administrative support from the Ministry of Labour, in which rotating cohorts of students were paid marginally more than they would otherwise earn on the dole. The explicit purpose was to combat unemployment and, by extension, the thriving gangs that cooked and sold methamphetamine. A visible and invaluable side effect, though, was that the state effectively subsidized a small cottage industry in woodcarving by also employing experienced carvers as trainees. The four carvers who have consistently worked there since the 1990s managed to do so, paradoxically, as students in the courses that they were teaching. I call them the pou (posts) who prop up the House of Knots. I worked there off and on for more than five years, carving daily for periods ranging from two weeks at first to multiple six-month stints in 2010 and 2011. From my very first week, in 2006, one of the carvers, whom I describe in more detail below, took me under his wing and taught me the techniques that comprise the craft – or at least as much as he deemed appropriate to my station, as a novice with no Māori ancestry (he didn’t care about that, he said, but some do. I’m from Lake Erie, in the USA, but I also grew up in Auckland). I loved the prayers and the songs, the wairuatanga (spirituality) and tohungatanga (expertise), and, most of all, the idle conversations around waist-high workbenches and on seats outside the roll-up garage door, where abalone half-shells served as ash trays. I also had an agenda, apart from learning to carve. In the name of ethnography, I was pursuing a speculative experiment that I intended to mirror the work done in big cities by professional Māori architects, who sometimes claim



‘Carving Costs Nothing’  77

that their practice is informed by Indigenous concepts of space and aesthetics that are traditionally held by carvers. From my place at the House of Knots, I was trying to imagine what urban planning might look like from the perspective of carving, through and against the grain of the real-world milieu in which architects actually work. More importantly, perhaps, my training in carving was afforded by field research grants, giving me the enviable opportunity to learn to carve without having to sustain myself by other means.1 In other words, I could carve for free because I did not have to labour for money – a privileged distinction that grouped me with a pair of elderly pensioners who joined us on a few big projects. That privilege motivates this chapter’s concern with the economies that sustain the House of Knots and its good works. Together, that pairing – i.e. to invent carving as a means by which to reflexively reinvent urban planning (Wagner 1981; Salmond 2017), and to be able to do so without participating in the jobs training course that enabled the House of Knots to exist – figured carving as a corpus of knowledge passed down from teacher to apprentice. From that perspective, carving is old yet variable across time and space, and genealogies of teachers remain intact as it evolves through different aesthetic movements, cultural revivals, government welfare programmes, repurposed workshops, and the like. The art form is comprised of patterns and the knowledge that one needs to be able to carve them. This includes not only techniques for holding and sharpening tools, but also chants, prayers, and ancient tapu (prohibitions) and tikanga. One should brush, rather than blow, the wood chips out of the way as they accumulate, for example, and gather wood chips and return them to the forest or the earth, rather than throw them in the rubbish or out of the back door. Indeed, the carvers in the House of Knots often talked about their vocation as tuku iho (handed down), albeit with two caveats: first, one needs to adopt a certain disposition of patience and reverence for this knowledge to take hold – for the corpus of carving to become embodied, as it were; second, like all Māori worlds, carving was devastated by settler colonialism but then also flourished through the natural course of innovation – a non-linear evolution that has left gaps in their expertise. They tended to fill these gaps by talking with other carvers and by consulting the collection of anthropology books on Māori art they kept in the shop’s kitchen.

The Carving and Kiwifruit Industrial Complex Over the three decades the four elder carvers – the pou – have worked there, the House of Knots has played host to a government-sponsored welfare-to-work programme. A cast of otherwise jobless men rotated through the workshop and earned marginally more than they would have collected

78  Jacob Culbertson



staying home on the couch. The less innocuous prospect was the lure of biker gangs, who sold drugs and catered to rather dark aspirations among disaffected men with little money and few means to get it. The course in Māori carving was one of several courses (with others in computing, horticulture, aquaculture and hospitality) that the local tribal trust board administered in conjunction with the Ministry of Labour and the New Zealand Qualifications Authority. The training programmes changed over the course of my time in the House of Knots, and this chapter traces those changes. But they were always framed by two objectives that complemented each other in some ways and conflicted in others: they offered vocational training and they were also the means to social development that would in principle scale up, such that a given training course is one node in a larger ecology in which Māoridom at large thrives and New Zealand becomes more bicultural. Thus, the courses are promising infrastructures in a double sense (Anand et al. 2018): both promising better lives in the future and built on the promise of greater self-determination as a long-standing vision for the relationship between iwi Māori (Māori tribes) and the state. Foremost, the courses were to address rural poverty (and by extension, reliance on state welfare) by growing an employable workforce. Upon completing a course, participants would earn a certificate attesting to their new set of skills and their general reliability in showing up to work on time and completing assigned tasks. With certificate in hand, the idea goes, they could get a job on a farm or in a small business. In reality, though, this solution to unemployment didn’t really match the problem. There wasn’t a lack of receptionists in town, no great demand for farming knowledge that couldn’t be learned on the farm, and for the duration of my time in those parts, the proposed mussel hatchery was just that: proposed. In other words, unemployment wasn’t so much a matter of competence (or much less, of certification) in entry-level tasks but rather inability to travel to nearby towns for work. When the costs of commuting would wipe out one’s minimum-wage margins, it wasn’t worth it. What there was in spades, however, were monocrop plantations of Monterey pines on the hillsides and, in the wide valleys below, the global centre of kiwifruit production. Those employers hired widely and frequently, and they would pick you up at your house in the morning. When hiring they looked primarily for soft skills – the demonstrated ability to show up to work on time, to work diligently and with a manner of respect for one’s co-workers, and to properly report workplace accidents (Bell 2017). All of the tribal training courses, regardless of their theme, furnished their graduates with certificates attesting to this capacity to conform to the norms of a corporate workplace, in addition to whatever specific acumen in computers or farming they may have learned as trainees.



‘Carving Costs Nothing’  79

The House of Knots occupied a unique place in this ecology of training courses and entry-level labour – what I facetiously call the Carving and Kiwifruit Industrial Complex – because it was the only course dedicated to a Māori kaupapa (a Māori project or programme).2 Its grounding in the tikanga of carving added value but it also put it in tension with the assessment criteria of a course designed to make carving useful for other ends, i.e. a general employability, as well as familiarity with the fundaments of Māori language and protocols. Ironically, in a course committed to hands-on woodcarving, the assigned carving tasks (usually pattern boards and masks with full facial tattoos) were supplemented by unit standard worksheets administered by the NZQA. Their purpose, obviously, is to enjoin the intimate, embodied craft of carving to a curriculum that can be evaluated and certified upon completion. Carving, in that vision, amounts to a set of procedures that smoothly translate between what one does with a chisel and the report on having done it. The actual woodcarvings – those material things made with tools and elbow grease under the instruction of caring, if terse, elders – become like worksheets, evidencing that one had moved through the curriculum, representing good work habits and the acquisition of knowledge. Both are whakairo in the frame of the NZQA curriculum, meaning woodcarving but also inscribing. As homework, the unit standards were met with apathy and resignation by mentors and trainees alike. Worksheets are poor substitutes for the education one gleans by attentively drawing out the figure that is already in the wood, as carvers imagine their craft (Grace 1986: 7). The requirement was particularly galling for the more advanced carvers. They passed their days in the shop – their putative training – making pieces to sell. This work of making carvings to sell also had the concomitant effect of making the physical and financial infrastructure of the shop (which was owned by the trust board and benefitted from funding from the Ministry of Labour) into a government subsidy for the local consumption of their wares. Without that chain of public funding there would not be a market for the trophies and birthday presents (and the occasional grand meeting house) that the carvers proffered in this cash-strapped local economy. The carvers would not be able to make ends meet otherwise. Filling out worksheets drew them away from their workbenches, to a table in the kitchen in order to report in worksheets on the task they had set aside. They invariably saw that routine as distracting from their real vocation and livelihood at the end of a chisel. What is more, the content of the unit standards (concerning patterns and tools, as well as research and documentation of one’s genealogy) was farcical for men who were elders in their communities, the keepers of knowledge that is guarded as sacred and not to be widely disseminated, including the more esoteric aspects of

80  Jacob Culbertson



carving. The effect was doubling, as a ‘figure seen twice’ (Riles 2002; Crook 2007), in which carving was both a corpus of knowledge (a curriculum of patterns and procedures to be learned, recorded and certified, and thus evidence that one can learn the patterns and procedures of any job) and also a set of relationships with materials, colleagues, students, and with the workshop, of whom they casually spoke as if it were a person. The carvers lived in this dissonance: the demands to demonstrate that they were learning about carving contravened what they understood— what they knew in their bones —carving to be.3 They were carving with strings attached. The House of Knots was tied to the job. Still, however distracting or annoying the busy work of the training curriculum may have been for these expert carvers, their role as tutor-trainees, tethered to their worksheets and cohorts of sleepy students, was not a ‘bullshit job’, David Graeber’s (2018) term for occupations that exist solely to pay a wage and occupy idle hands – far from it. These carvers knew the men who came through the training programme (or knew who they were, anyway), and they honoured them as a matter of course, not just with the simple respect for another person, but also as the embodiment of their family lines and as corporeal extensions of their ancestral lands. Up and down this stretch of coast, everybody is related to somebody’s relatives. Some of the shop’s patrons had it bad— being bipolar, re-entering from prison or tempted by gangs and drugs. Others appeared drawn to the shop by the ‘traditional thing’ (Povinelli 2002), the hopeful figure of culture (and art in particular) as a bridge between pre-colonial pasts and futures or, more pessimistically, the root of the cruel optimism (Berlant 2011) that the kaupapa Māori in this state-sponsored training course might offer them an alternative to the settler-colonial capitalist ‘regime of living’ that pathologized them as unemployed and underdeveloped (Collier and Lakoff 2007: 22). In any case, knowing the relations and honouring the living faces of ancestral lines that passed through their workshop joined the carvers at the House of Knots in a matrix of mana whenua (power from the land), a cat’s cradle that constantly recomposed the geography of kin beyond their front gate. Conspicuous hospitality (manaakitanga) makes worlds in Te Ao Māori, the Māori world. In the House of Knots, that important ethic and practice was contingent upon the young men that their training courses brought through the shop. This relational worlding is important work in its own right, a raison d’être, and it is hardly reducible to jobs training. The carvers’ work was not driven by an investment in some Māori ideals either, much less by altruistic charity. They would hardly fit the bill of sacred guardians of sacred Māori arts, who impart their basket of carving knowledge to those who prove worthy. Yet, they certainly stuck to their tikanga (no food in the workshop, keeping the workspace clean, saying prayers)



‘Carving Costs Nothing’  81

and taught those who had the patience to learn. No one was turned away. My teacher always said that anyone who can listen can carve, though not everyone can slow their minds and hands enough to listen long enough before getting bored or frustrated and drifting off. Just being around that careful work, and its requisite patience and reverence, can make a man stand a little taller. But the carvers were much too irreverent, and simply too committed to bringing forward the ancestors in the wood, to claim their awkward mantle as the tohunga (experts) of workplace norms and social welfare providers to the down-and-out, building the worlds envisioned by TribalTrustCorp and rubber stamped by the NZQA.

Relational Space Besides The Benefit, as the welfare dole is called, the benefit to them was that the course brought people through the door. People coming through the door is integral to the philosophy and technique of carving in the House of Knots. From my earliest days there, I was struck by how they would invoke the spatial principles of marae (meeting grounds) and wharenui (meeting houses). Those iconic forms are widely regarded as architectural objectifications of indigenous modes of kinship and sociality (Salmond 2009; Neich 2002; Sissons 2010). Less noted, though, is that spatial forms pervade more colloquial ways of talking about and enacting space as relational rather than absolute (Ka‘ili 2017; Māhina 2010,). Invoking, enacting or activating a transformative space of encounter and exchange is the desired effect of the ritual encounters on marae that make these fundamental principles visible. Fluency in deploying incisive analogies in whaikorero (formal exchange of speeches) is an expert mode of enacting relationships in its own right (Biggs 2006). When deployed in casual conversation, as ‘metaphors we live by’ (Lakoff and Johnson 1980), they can conjure images of encounter, and their animating life-force, in everyday spaces. For example, when I first arrived at the House of Knots, early one summer, a late spring cleaning was about to commence: sweeping the concrete courtyard, hosing out sawdust, and pulling weeds from the cracks between the slabs. The regular carvers sat to rest on crude benches just outside the roll-up shop door, presiding over their domain – like an ātea (open area) on the crumbling asphalt of an old shoe factory. My teacher told me that when the conversation got good, and when we laughed and sang to bide the long days on the carving floor, our words went into the bones of the house, into its backbone, and propped it up. It made it warmer. The allusion is to houses as bodies (meeting houses have human anatomies, each part with a function) in a world in which bodies are also made of stories about ancestors who occupied houses before.

82  Jacob Culbertson



I once saw my teacher in a rare burst of anger yell to another carver to stop chasing a bird that had flown in through a window, because the bird had only come to help. He winked at me when praying mantises landed on my hat – three times in as many days, in a week during which a big project pushed us all to work into the night. He said the wairua (spirit) of the place was thicker and more present when the shop filled with the chorus of tapping mallets. These were passing moments. Perhaps those words were just rhetorical flares. But they unsettle the notion that buildings are just containers for what happens in them. They dispatch with the figure of industrious men making culture in a material world, with tools and houses at their disposal. The House of Knots was creativity writ large, an ecology of practices (Stengers 2005), at the centre of its universe.

A Love Story My teacher was the oldest among the regular carvers. He was a retired Telecom lineman who had come to the shop after he slipped in a hunting accident and landed on the permanent disability benefit. As he tells it, he broke his hip late in the day, too late to hobble out, so he strapped his two dogs to his body with his belt to stay warm through the night. Some days later, when the doctor told him to stay a while in the hospital, he replied that his dogs were getting too fat and that good dogs need to run. So, he went home. He has walked with a limp for as long as I have known him. The benefit is just enough to live, enough to carve, with a little left over for a few handles of amber beer on Thursdays. I call him Matua,4 a term of respect for an older man. We met in 2006, in the House of Knots, when a friend parked me there to pass the day while he went to work. I spent that first day, and then the second, peering awkwardly from the far side of the workbenches at these people whom I had just met, watching hands and chisels. On the third day, Matua handed me a chisel – a whao haehae for v-shaped lines – and demonstrated the two cuts that turn curvy triangles into the three-toed foot of the pukeko bird, spaced regularly in a pūngāwerewere (spiderweb) design. My apprenticeship (for lack of a better word) would proceed that way for years. He never told me much about carving. He showed me. There was no theory, only lines and cuts and later songs and prayers, the meanings of which he mostly left to me to discern. I learned to interpret the reaction that he expressed in his eyebrows – raised for ‘could be right’, furrowed for ‘a bridge too far’ – when I asked him to indulge my guesses. He would suck his teeth and roll his eyes if I my expositions tended toward the fanciful. If I asked him to translate a word in Māori he would say, ‘there’s a dictionary in the kitchen’. Once, for a period of a month or so, I had to go outside if I



‘Carving Costs Nothing’  83

wanted to speak English, a rule that applied to no one else. He told me that he learned a lot as a kid by climbing atop another kid with broad shoulders and peering into the windows of the meeting house to eavesdrop on the affairs of the old people. Knowledge was piecemeal and literally required some stretching. He used the metaphor of light passing through a tightly gapped picket fence, composing a bigger picture like constellations in the night sky. On Thursdays, when the benefit checks came in, we went to the pub and did the crossword puzzle. That was Matua’s regular practice and it became mine over the years. It had a sequence to it. We would talk to each other and to the bartender over the crossword puzzle. We never talked about carving in the pub. Then, in the late afternoon, a cast of regulars – a pair of fence-builders, a dairy farmer, the high school principal, and a helicopter pilot – would assume the same stools at the bar they had already thirty years prior. We drank faster with them, because we talked a lot and it was louder. Then, around five beers and five o’clock, Matua would start shaking people’s hands with tipsy affection. Matua had an aphorism that he would recite when the mood was right. ‘Aroha costs nothing’, he said. ‘Look after it’. Aroha is usually translated as love, but it also more than that. Aro means form or appearance and, as a verb, to pay attention or to face towards something. It is the shape of things. Hā is breath, a substance at once particular and universal, and a medium of exchange that conjoins people most visibly when they greet each other by pressing together their foreheads and circulating the air of one nose through the passages of another.5 It is also taste, or sensuousness as a medium. More than love, the etymology of which connotes desire and attachment, aroha is a realm of knowledge and action that is concerned with what is perceptible to the senses, with becoming entangled rather than being possessive. Whakarongo (to listen) entails a receptive, calm disposition, to fill one’s basket of another, to nourish and cultivate peace. But these words are not enough. Look at your right hand. Palm up. Your little finger, the pinkie: te ihi. The power. Ringer finger: te wehi, that which makes the hair on your neck stand up. Now the longest finger, the one that stands the tallest: te mana, the authority! Index finger: te mauri, the animate life-force. The thumb: te wairua, the ‘Holy Spirit’ in the Bible, the water in which we fish swim.

Matua says, ‘all of these are given, they are not for you to hold. But the aroha’, he closes your fingers into a fist around your palm, ‘the aroha costs nothing’.

84  Jacob Culbertson



Te ihi, te wehi, te mana, te mauri, te wairua. The aroha stands alone. I don’t know where that comes from, and I wouldn’t presume to ask. Paul Tapsell (1997: 330), Te Arawa anthropologist and expert in the practices of caring for taonga (treasures) and their genealogies, names the third term in the sequence as wana, which he defines as ‘authority; class; integrity; unquestioned competence’. The scheme is an indigenous philosophy of aesthetics, applicable to the power of artefacts to hold the presence of revered ancestors and to take on the so-called magical qualities of ancestral agency that Alfred Gell (1999) called ‘technologies of enchantment’. But it is also a practical guide to encounters of all kinds, of how to find your way through a trying situation. It describes the steps through which one moves when coming on to another’s meeting grounds, when facing the carved face of a meeting house, reaching out with its fastest warrior whose face may also be carved. Meeting houses reach out to their visitors, who meet them halfway. Aroha is a way of training your attention, of slowing it down, to listen carefully. I draw words from an academic ancestor of my own to picture it more clearly: imagine the paddles of an electric fan – ‘run at high speed, [they] appear to be one continuous, whole, circle of matter, but when you slow them down, it may no longer look like one continuous entity’ (Klima 2002: 213). In retrospect, it seems that Matua would pull out this trick – he would hold my right hand – whenever our idle talk would drift towards worries or anxieties or proud boasts or snide jokes. Aroha is the middle path, the attentive equanimity to receive things as they are, to be receptive to the figure that is already in the world. The flow of men through the House of Knots – those who came for a cup of tea, a jobs training certificate, or to work in good company on a renowned project6 – was not separate from their creativity. They were integral to it. No one was in the way.

Community Max I spent 2009 across the ocean in California and returned to the House of Knots at the start of the following year. On arrival, I found that they were no longer running the jobs training course. It had been jettisoned by the reigning centre-right National Party government, amid a relatively mild gesture to austerity in the wake of the 2008 economic downtown. Instead, the House of Knots was now running a new, different welfare-to-work scheme, called Community Max. This new government programme was based on the hope that a quick injection of cash into poor small towns may knock beneficiaries off the welfare rolls for good. To that end, the programme interfaced with the Work and Income New Zealand, which aids and monitors the efforts of their constituents to find jobs, as a condition of



‘Carving Costs Nothing’  85

their welfare benefits. Community Max resembled the old training course in that it sought to incentivize the unemployed to seek practical work experience that they could list on a resume. Participants were also awarded a certificate of completion. However, Community Max was crucially different in that it was temporary and focused on small, short-term, one-off community development projects, identified in cooperation with civic leaders. The programme was nationwide in scope but explicitly addressed youth unemployment in predominantly Māori regions (New Zealand National Party 2010). In fact, half the participants in that programme were Māori and numerous projects operated on or near marae. For example, one hapū (sub-tribe) close to the House of Knots cleared bush and forged a track to the top of their sacred mountain. Another excavated and rebuilt a pre-colonial pā (fortified village) to its original scale, building bridges and small houses out of the trunks of tree ferns. Another group of carvers on the opposite side of town refurbished the carvings from a long-shuttered meeting house that had been built by a prophet of great renown more than a century earlier. For their part, in that summer of 2010, the House of Knots pitched in on a shade structure for the Māori King, who was expected for the Kapa Haka Festival, a competition of cultural performance. Six young men under the guidance of an expert boat builder lashed a post-and-beam structure with ancient maritime knots and covered a roof of chicken wire and black plastic sheeting with palm thatch. However inspiring, those projects with their changed funding structure brought a tinge of anxiety to the House of Knots. While a couple of the carvers were state beneficiaries for life due to various health reasons, the other two foresaw the need to return to the assembly line in the kiwifruit packhouse or head up to the bush on a forestry crew. They had the skills and experience to slide into jobs with more pay and more responsibility, certainly enough to feed their families. But it would take them out of the House of Knots, at least seasonally. Nevertheless, the overall mood was one of cautious optimism: the dissolution of the permanent training course and its short-term, occasional replacements could also be an opportunity for financial and artistic independence. They wanted to carve for an art market, which they imagined could fill the role that the government funds had once performed in subsidizing their provisions of small taonga that their neighbours buy as gifts to hang in living rooms and bring to retirement parties. In that vein, in the downtime between cohorts of Community Max participants, we embarked on an experiment that was surreal in the beginning and doomed in the end. One of the carvers had been contacted by a woman in Australia who wanted to leverage the Mozzy (Māori in Aussie)

86  Jacob Culbertson



demand for twenty-first keys (a common gift for twenty-first birthdays that is literally the shape of a giant door key with carving patterns and often a photo embedded in the round end). Apparently somewhere in Australia cheap and inauthentic carvings made on factory machines in China can be had at cut rates in souvenir stores, as is ubiquitous in tourist traps throughout New Zealand. Handmade twenty-first keys thus have the added value of authenticity, and even more so when they carry the story of a small-town workshop. So, over the course of a week, the House of Knots set up a Fordist assembly line, complete with targets for how long it should take to draw a key, cut it on a bandsaw, carve a pattern, glue the abalone eyes, and so forth. In the end, it proved impossible to make the investment of time match the return on the money, as figured by shipping costs and bulk pricing. If nothing else, the experiment confirmed the precarity of real carving in an age of mechanical reproducibility (Benjamin 1968) and the importance of public funding to the worlds of not-quite-art-markets that are strung together by the travels of everyday, modern-day taonga. That subsidy collapses a tidy distinction between gifts and commodities and allows the fruits of those carvers’ labours to be both a source of petty cash and affective icons of ancestral presence in the everyday lives of everyday people. The big project over the ensuing months was to carve poupou (ancestor panels) for a meeting house up the coast. The carvers were paid in cash, which they supplemented with income from selling small taonga, which was in turn subsidized by Community Max projects that paid the trust board the money needed to cover the utilities and taxes on the old shoe factory that housed their ambitions. I wandered off to Auckland after that, to chase architects for six months, and returned to the House of Knots in the summer of 2011. I arrived to find the carvers excited about a project to make dozens of carved posts for a new cycle track that would stretch for two hundred kilometres along the beach and over the bush-clad mountains before reaching a different coast. The central government had initiated a series of cycle tracks to bolster tourism economies in remote areas and had invited local and district councils to submit budgets. This particular cycle track was a collaboration between two neighbouring district councils and the Department of Conservation. The route of the track would pass through and conjoin the territories of three neighbouring Māori tribes, whose tribal governments were also invited to participate in the design and construction of the cycle track. The plans for the track included informational signposts along the route that would detail the local heritage and history, with carved bases supporting them. Large carved gateways, carved by those from the participating tribes according to their various local styles, would mark either end of the cycle track.



‘Carving Costs Nothing’  87

The opportunity to carve totem-like posts for the cycle track was a source of considerable excitement around the carving shop. Initially the carvers discussed it as an opportunity to bring together the best carvers in the area under one roof. Just as the route of the cycle track would link together the mountains and valleys where those carvers lived, the collaboration would reiterate the geography of historical relationships among neighbouring tribes. Each would carve the posts that would stand in proximity to the places where they lived and worked, making visible a geography of contemporary carvers that would be contained, silently, in the posts that mark points of historical significance or route-finding for cyclists. While such a network would have no centre, the possibility that it would emerge from their modest carving shop in town was the source of pride and excitement. Hosting such conspicuous events would contribute to the standing of their workshop, in a dual sense – such activities would add prestige but also, according to the carvers, the talk, prayers, laughter and sweat that move through the shop also permeate the walls of the building and strengthen them. More than just promising economic benefits, the cycle track offered a new opportunity for tribal elders to educate the community about the landscape that the track would pass through. One brisk fall day, a dozen of us piled into pickup trucks and spent the day driving the route of the impending cycle track, tracing it as closely as we could with the existing roads. Along the way, one prominent elder and spokesman recounted the names and histories of the peaks and streams that we saw. Members of the group had memories of their own, mostly big fish stories of good hunting spots deep in the bush or recounting memories of the farms and villages that had long-since vanished into the landscape as people moved into town or to the cities. At one point we came across a group of trail-builders from Auckland who had with them a fleet of utes and grading machinery on trailers, high in the bush. Our group was polite and exchanged small talk, but when we climbed out of our own fleet of trucks at the next stop, the conversation had a tone of disbelief and frustration. If there is one asset that this relatively depressed rural economy has in spades, it is farmers and trail builders. People here know how to get around in the bush and how to build a trail. Yet the cycle track – which promised to bolster the local economy by creating jobs through tourism – was being built by private contractors from Auckland. The following day, back at the carving shop, a representative from the tribal trust board came by to discuss the plans for the carvings that would line the cycle track. He explained that he had procured a Community Max scheme to fund the work. We knew from experience that this funding would be distributed as wages. The scheme allowed for one paid tutor, while all other recipients must already be on the dole to be eligible, because, like all Community Max programmes, the point was not to

88  Jacob Culbertson



produce woodcarvings or even cycle tracks, but rather to get young men off the dole. In this case, then, there would be no available funds to bring together working expert woodcarvers. Even three of the four carvers in the shop would not be paid at all. Feeling provocative, I asked why the tribal trust board was paying outside contractors to build the cycle track while only one of the four regular carvers could be paid for the work that he would do, which now included mentoring young men while also doing the actual carving. The tribal trust board representative’s answer was striking. He said, ‘We are able to pay for skilled work like building a cycle track. But for carvings we need to run it as a jobs course’. It was an off the cuff comment and a slip of the tongue, no doubt. But it highlights the way in which the House of Knots was charged with also being so much more than a carving shop. The various job-creation schemes that have run through that place are valuable precisely because they leverage the power of kaupapa Māori and the accompanying imaginaries of learning and performing uniquely Māori ways of being – practices that make their ancestors appear. But by the logics by which the state funds carving, Māori people doing even the most sacred and rarefied things is putatively natural. Avenues for mobility, however, must be adequately resourced and expertly built.

Carving for Kai The first time I met Matua, he told me ‘we only carve for kai [food]’, an ‘empowering refrain’ (Stengers 2008) that underwrites his ethical commitment to his own brand of gumboots-and-overalls asceticism. Initially, I thought he meant that people brought them food to keep them carving – a reasonable assumption in those late summer months, when the kahawai fish spawn so thick that they jump onto the river banks and the fishermen share their fresh glut with the elders, including the carvers. The House of Knots ate fish stew for weeks. So of course, I was incredulous when, over those same hot weeks in 2006, I watched Matua greet his friends and his friends of friends at the House of Knots’ chain-link gate and hand over twenty-first keys with one hand and take thin envelopes of cash with the other. Now, after so many years, I can imagine that ‘only carving for kai’ meant that the price of their carvings is inseparable from who is buying them and what that person can pay. They don’t haggle. The cost of their carvings might be better understood as a donation. Perhaps he also meant that they cannot carve to enrich themselves because they cannot access an art market in which carvings are commodities. Instead, in those early days, the House of Knots made ends meet by running a carving course that in turn subsidized carvings for cash-strapped locals, the income from



‘Carving Costs Nothing’  89

which passed through the coffers of the state to ultimately put food on their tables. Indeed, in this roundabout way, they carved for food, but not the food that was delivered to the shop. In retrospect my assumptions about those exchanges, and the aphorisms with which Matua explained them, count among many subsequent uncontrolled equivocations – instances in which what I heard and understood was not what he meant (Viveiros de Castro 2004). We were inventing each other, per Roy Wagner’s (1981) conception of anthropology’s vocation. When I arrived at the House of Knots, knowing little about carving, I was motivated by a research project that posited a symmetry between traditional carving and more modern modes of professional Māori architecture. That proposal departed from a notion that carving was a different practice, belonging to a different economy, one that was by definition outside of (and prior to) a colonized world in which the modern forms of state-craft (including in architecture and economics) are more universal. In other words, by this light, carving was indigenous while architecture needed to be indigenized. Conversely, Matua was also inventing me as his student, with an education in carving that demanded that I try to see from his perspective, speculatively and imaginatively; or better put, to understand carving and the world it makes by seeing it through his tutelage and his aphorisms – perhaps a more robust notion of becoming an apprentice than the apparent concern with the correct techniques to hone a chisel’s blade or the rote memorization of waiata koroua (old man songs). In some obvious sense the House of Knots does indeed participate in two economies, practising two kaupapa simultaneously, or enacting two relational ecologies. On one hand, it is a node in a larger institutional ecology of jobs training courses that conjoin state welfare reform, tribal trust boards and agricultural wage economies. The art of carving is one context in which this technocratic assemblage is forged and maintained – what I call the Carving and Kiwifruit Industrial Complex. On the other hand, the House of Knots is also an extension of ancient indigenous modes of creativity and expertise, for which a welfare-to-work scheme in an old shoe factory is one of many, diverse sites in which carving thrives and innovates. Amidst the bureaucracy of worksheets and funding that comes and goes, the work of carving also maintains networks of carvers and cousins, and principles of manaakitanga and whanaungatanga (familiality), in ways that are unrelated (and apparently invisible) to the regime of workplace qualifications that govern the jobs courses. In other words, it would be appropriate to count the House of Knots as another instance of indigenous art moving through different regimes of value (Myers 2002; Michaels 1994), or of vernacular craftsmen maintaining their traditions (Sennet 2009; Marchand 2009). That conceptual

90  Jacob Culbertson



convention of parsing their practices from the context in which they actually take place (i.e. as old venerated protocols that adapt to novel modern institutions) certainly resonates with academic narratives in New Zealand, of encounters between two worlds, each with different governing logics (Salmond 2014). In those terms, the dual economy of the House of Knots exemplifies the encounter between Te Ao Tawhito (the old world) and Te Ao Hou (the new world), or Te Ao Māori (the Māori world) and Te Ao Pākehā (the settler world). We could perhaps even parse the House of Knots into two indigenous concepts of exchange, hoko (mundane barter) and tuku (momentous gifting) (Salmond 2012). The distinction is useful to accentuate how the commodity relations that make people into things by conforming them to the ‘abstract labour’ of the assembly line also makes things (i.e. ancestral carvings) that are (proxies for) people (Henare et al. 2007; Gell 1998; Strathern 1990). That is, the carvings that job trainees sell for money still continue to proliferate the presence of ancestral worlds in everyday lives and households and do so precisely by travelling through whakapapa, the indigenous webs of kinship and relationality through which people already know each other. Still, however useful as heuristic concepts, these binaries may obscure more complex dynamics and emergent modes of indigenous relationality and practices of space-making. For one thing, the history of contemporary Māori carving is deeply entangled with the infrastructures of the state. In fact, in some ways, carving itself is an infrastructure of contemporary bicultural New Zealand. As an object of deliberate, self-conscious cultural revival since the mid-twentieth century, carving has long figured in topdown efforts to revitalize Māori economies, especially in the work of the state-supported New Zealand Māori Arts and Crafts Institute (NZMACI) to ‘traditionalize’ carved meeting houses (Sissons 1998; Lythberg and McCarthy 2019). In fact, the carvers of the House of Knots descend directly from this lineage, having initially practised under a tutor who attained his craft (and his credentials) from the NZMACI. They were able to merge carving with social and economic development because those seemingly disparate fields of practice had already been entangled with each other for decades before the House of Knots first opened its doors. More importantly, though, the conventional imaginary of indigenous traditions encountering (and being appropriated by) the infrastructures of the state may also elide what was so unique about the House of Knots, for whom carving was explicitly a practice of aroha. In 2010, when the carvers scrambled to make ends meet between short-term Community Max projects, there was also a corresponding boom in social welfare schemes targetting Māori men on the dole. No doubt, those schemes save lives and find jobs, but some of their constituents fall through cracks. That



‘Carving Costs Nothing’  91

is who kept coming to the House of Knots when the jobs training dried up, to bum a cigarette and a coffee and to sit with old men doing visibly Māori things. They were drawn to the ‘lifeforces’ in the shop, Matua said, and their presence would feed back into the carvings, which in turn would go into their families’ homes and meeting houses – places where many of these young men didn’t feel comfortable amid the formalities. Perhaps it is this notion that carvings are born from the lives that passed through the workshop, as if the House of Knots was metabolizing them, that can challenge and enrich conventional ideas of how architectural spaces shape subjectivities, and thus also how indigenous building practices are selectively appropriated by contemporary architectural practices as sources of formal and ethical inspiration. The House of Knots, after all, operated at the interface of two different ways of making persons through the work of making carvings – a symmetry that makes the difference between these regimes of value even more pronounced. By what measure could the House of Knots be recognized for making a space in which Māoritanga (‘Māoriness’) thrives, in catering to those who are invisible to state welfare programmes and arguably even to the formal strictures of sacred meeting grounds, and yet still embody their houses and make those institutions stand? No question, the House of Knots has served as an extension of a state ‘biopolitics’ over those decades – a site of a programmatic, pastoral concern for ‘life itself ’ among depressed men in a depressed rural economy (Foucault 2003). More than just subsidizing the local economy of carving, gathering people to their open door and forecourt was also integral to what House of Knots is and what they make – a dynamic composition of such creative forces as elderly pensioners telling old stories and old jokes, the teenage misfits pulling a paycheck, the birds and the breezes, that tingly feeling that something is up when an ancestor carving is almost done, and the idle talk that goes in the bones of the house and keeps the walls warm. The House of Knots has no truck with the Enlightenment figure of the solitary creative individual, exerting his will on inert materials. And they do not carve for money; not really. Rather, theirs is a relational practice of spatial production, affected through mundane rituals in the workshop – of biding the time with games of trivia, of recounting the gossip over rolled cigarettes, of gathering to make hip-hop beats on a laptop in the kitchen, and, in the hot days at the end of the summer, of one pot after another of kahawai fish stew – and through the travels of carvings that retrace and reweave the genealogical ties that bind extended families to each other and to their ancestral lands. Yet if one holds too narrowly to the notion that carving is outside of and prior to the economic development programmes of the state – as if carving is essentially an ancient practice, innovated and appropriated through its

92  Jacob Culbertson



history7 and available to selectively inspire new architectural practices under the banner of indigenous knowledge or technology – that perspective may make it impossible to see what carving actually is in this particular place. It is that which reaches beyond the disciplinary infrastructures of the state, emerging through those infrastructures as a relational practice of care that resists measure, harnessing and trafficking in lifeforces of a different sort, at the centre of its universe, in an old shoe factory. Jacob Culbertson is Senior Design Strategist in Workplace and Technology at the global design firm Gensler in San Jose, California. He earned his PhD in Sociocultural Anthropology from the University of California, Davis in 2015. He has held academic appointments at UC Davis and Haverford College, a Postdoctoral Fellowship with the Wenner-Gren Foundation, and an multi-year apprenticeship in Māori woodcarving at the House of Knots.

Notes   1. The research for this chapter was supported by The Wenner-Gren Foundation, The Graham Foundation for Advanced Studies in the Fine Arts, and the University of California’s Pacific Rim Research Program.  2. Kaupapa Māori is a field of vibrant innovation in ‘indigenizing’ or ‘decolonizing’ modern institutions, for intervening in educational institutions in particular (cf. Smith 1999, for example). While that literature may indeed be relevant to this chapter, my use of the term kaupapa is not intended to signal or engage with that conversation. Rather, I am using its vernacular meaning, as a project or programme.   3. The carvers themselves made this distinction between modern and ancient conventions of teaching their craft, for example, when they would remind me, wistfully and frequently, that in the old days an apprentice would have to bite the master’s toilet seat.   4. I have chosen not to identify Matua by his personal name, in keeping with ethnographic conventions of using pseudonyms.   5. This is my own explication, based on my experiences in the Māori world and my familiarity with Te Reo Māori (the Māori language). It is neither a dictionary definition nor a recapitulation of meanings that I have been expressly taught by any other person.   6. For example, the two grandfathers who came daily for three months in 2006, when we carved architectural pieces for a museum installation. Or the boat builders who helped assemble a shade house for the Māori King. Or me, a fledgling American anthropologist. Big, venerated projects tended to attract people who did not regularly work there otherwise.



‘Carving Costs Nothing’  93

  7. For example, one early reviewer of this chapter suggested that carvers are not concerned with money and their anthropologist friends should not write about their political economy because it’s not actually part of what they do. Here I am suggesting that, in the House of Knots, the relation between creativity and biopolitics is ‘integrally inferred’ (Wagner 1991).

References Anand, Nikhil, Akhil Gupta and Hannah Appel (eds). 2018. The Promise of Infrastructure. Durham, NC: Duke University Press. Bell, Lindsay. 2017. ‘Soft Skills, Hard Rocks: Making Diamonds Ethical in Canada’s Northwest Territories’, FOCAAL: Journal of Global and Historical Anthropology 79: 74–88. Benjamin, Walter. 1968. Illuminations. New York: Schocken Books. Berlant, Lauren. 2011. Cruel Optimism. Durham, NC: Duke University Press. Biggs, Bruce. 2006. Kimihia te Mea Ngaro: Seek That Which Is Lost. Auckland: Polynesian Society; Macmillan Brown Centre for Pacific Studies. Collier, Stephen J. and Andrew Lakoff. 2007. ‘On Regimes of Living’, in Aihwa Ong and Stephen J. Collier (eds), Global Assemblages: Technology, Politics and Ethics as Anthropological Problems. New York: Blackwell. Crook, Tony. 2007. ‘Figures Seen Twice: Riles, the Modern Knower and Forms of Knowledge’, in Mark Harris (ed.), Ways of Knowing: Anthropological Approaches to Crafting Experience and Knowledge. New York: Berghahn Books. de la Cadena, Marisol and Mario Blaser (eds). 2018. A World of Many Worlds. Durham, NC: Duke University Press. Foucault, Michel. 2003. ‘Lecture 11, 17 March 1976’, Society Must Be Defended: Lectures at the College de France. New York: Picador Press. Gell, Alfred. 1998. Art and Agency: An Anthropological Theory. Oxford: Oxford University Press. . 1999. ‘The Technology of Enchantment and The Enchantment of Technology’, in The Art of Anthropology. New York: Berg. Grace, Patricia. 1986. Potiki. Auckland: Penguin Books N.Z. Graeber, David. 2018. Bullshit Jobs: A Theory. New York: Simon and Schuster. Henare, Amiria, Martin Holbraad and Sari Wastell (eds). 2007. Thinking Through Things: Theorizing Artefacts Ethnographically. New York: Routledge. Ka‘ili, Tēvita. O. 2017. Marking Indigeneity: The Tongan Art of Sociospatial Relations. Tucson: University of Arizona Press. Klima, Alan. 2002. The Funeral Casino: Meditation, Massacre, and Exchange with the Dead in Thailand. Princeton, NJ: Princeton University Press. Lakoff, George and Mark Johnson. 1980. Metaphors We Live By. Chicago: University of Chicago Press.

94  Jacob Culbertson



Latour, Bruno. 1993. We Have Never Been Modern. Cambridge, MA: Harvard University Press. Leach, James. 2004. Transactions and Creations: Property Debates and the Stimulus of Melanesia. New York: Berghahn Books. Lythberg, Billie and Conal McCarthy. 2019. ‘Te Ao Hou: Whakapapa as Practical Ontology’, The Journal of the Polynesian Society Special Issue 128(1). Māhina, ‘Okusitino. 2010. ‘Tā, Vā, and Moana: Temporality, Spatiality, and Indigeneity’, Pacific Studies 33(2/3): 168–202. Marchand, Trevor. 2009. The Masons of Djenne. Bloomington: Indiana University Press. Michaels, Eric. 1994. Bad Aboriginal Art: Tradition, Media, and Technological Horizons. Minneapolis: University of Minnesota Press. Myers, Fred. 2002. Painting Culture: The Making of an Aboriginal High Art. Durham, NC: Duke University Press. Neich, Roger. 2002. Painted Histories: Early Māori Figurative Painting. Auckland: Auckland University Press. New Zealand National Party. 2010. Press release: ‘Community Max programme expanded’. 24 June. Scoop Independent News. Retrieved 5 March 2020 from https://www.scoop.co.nz/stories/PA1006/S00384.htm. Povinelli, Elizabeth. 2002. The Cunning of Recognition: Indigenous Alterities and the Making of Australian Multiculturalism. Durham, NC: Duke University Press. Riles, Annelise. 2002. The Network Inside Out. Ann Arbor: University of Michigan Press. Royal, Te Ahukaramū Charles (ed.). 2003 The Woven Universe: Selected Writings of Rev. Māori Marsden. Otaki: Estate of the Rev. Māori Marsden. Salmond, Amiria. 2005. ‘Nga Aho Tipuna: Ancestral Threads’, in Susanne Küchler and Daniel Miller (eds), Clothing as Material Culture. New York: Berg. . 2017. ‘Uncommon Things’, Anthropologica 59(2): 251–66. Salmond, Anne. 2009. Hui: A Study of Māori Ceremonial Gatherings. Auckland: Raupo Publishing. . 2012. ‘Ontological Quarrels: Indigeneity, Exclusion and Citizenship in a Relational World’, Anthropological Theory 12(2): 115–41. . 2014. ‘Tears of Rangi: Water, Power, and People in New Zealand’, Hau: Journal of Ethnographic Theory 4(3): 285–309. Sennett, Richard. 2009. The Craftsman. New Haven, CT: Yale University Press. Sissons, Jeffrey. 1998. ‘The Traditionalisation of the Maori Meeting House’, Oceania 69(1): 36–46. . 2010. ‘Building a House Society: The Reorganization of Māori Communities Around Meeting Houses’, Journal of the Royal Anthropological Institute 16(2): 372‒86. Smith, Linda Tuhiwai. 1999. Decolonizing Methodologies: Research and Indigenous Peoples. London: Zed Books.



‘Carving Costs Nothing’  95

Stengers, Isabelle. 2005. ‘Introductory Notes on an Ecology of Practices’, Cultural Studies Review 11(1): 183‒96. . 2008. ‘Experimenting with Refrains: Subjectivity and the Challenge of Escaping Modern Dualisms’, Subjectivity 22(1): 38–59. Strathern, Marilyn. 1990. The Gender of the Gift. Berkeley: University of California Press. Tapsell, Paul. 1997. ‘The Flight of Pareraututu: An Investigation of Taonga from a Tribal Perspective’, The Journal of the Polynesian Society 106(4): 323–74. Viveiros de Castro, Eduardo. 2004. ‘Perspectival Anthropology and the Method of Controlled Equivocation’, Tipiti: Journal of the Society for the Anthropology of Lowland South America 2(1): 3–22. Wagner, Roy. 1981. The Invention of Culture. Chicago: The University of Chicago Press. . 1991. ‘The Fractal Person’, in M. Strathern and M. Godelier (eds), Big Men and Great Men: Personifications of Power in Melanesia. Cambridge: Cambridge University Press.

6

Zombie Architecture Sacrifice in Pre-contact Polynesian and Classical European Buildings ♦l♦

Ross Jenner and Albert L. Refiti

For, the object of the sacrifice precisely is to establish a relation, not of resemblance, but of contiguity, by means of a series of successive identifications. —Claude Lévi-Strauss, The Savage Mind

T

his chapter juxtaposes two architectural traditions to speculate upon the roots of both cultures in terms of sacrifice, a motive no longer relevant to architectural production in either, given that Christianity put a stop to blood sacrifice and generally diminished the importance of sacrifice in both cultures. Hence, we propose an architectural dialogue between Pacific and Western ideas concerning the role of ritual and sacrifice in structuring the origins of sacred buildings of the West and the Pacific through notions of the (re)animated dead. This dialogue, in which we position Pacific sacred buildings alongside their European equivalents, is important because we want to see if there is a common ground: firstly, in the way sacrifice is involved in the sanctification of buildings; secondly, in the way language tropes are used to describe spaces and architectural ornament linked to sacrifice; thirdly, the use of architectural forms such as mounds and altars as central to rituals of sacrifice; and, finally, the use of



Zombie Architecture  97

ornaments in and on buildings as the ‘concretion or the reconstruction of the dead person within’ (Hersey 1989: 20). We term this situation ‘Zombie architecture’, denoting attempts to embody the living dead, which take on significant roles as ornament and ritual attractors within buildings (Kahn 2008) – they reanimate the dead within the space of the living. The wooden altar frame of Greek architecture, which George Hersey speculates to be the origin of the temple, was used to hang sacrificial body parts that were dismembered on a table. Our conjecture is that something similar may be found in Pacific buildings. Reconstructing the ground by mounding, heaping or carving of earth and stone binds and fixes the dead ‘in place’. There is also the link between blood sacrifice and sanctification, which encloses these sites as tapu (sacred, set apart, restricted) within a temenos (sacred precinct). Both Classical European and Pacific architecture are redolent with images of sacrifice. The Parthenon in Athens is literally encrusted with images of blood offering.1 The Hawaiian luakini heiau (sacrificial temple) was principally used for human sacrifices: ‘when he goes to war, the king offers a human sacrifice at the luakini temple of Kū, but at other times he consecrates temples to the agricultural gods Lono and Kāne’ (Kirch 2010: 41). Our aim is not to recount all the examples of sacrifice in Pacific and European architecture, rather, to explore the idea that real events organized around ritual killing were necessary to the origin of buildings in both traditions. The arrangements of the rituals were the formal and spatial means to accommodate the performance of such events, which gave rise to primary structural elements of the architecture. Thus, we argue, sacred buildings in European and Pacific architectural traditions are connected to sacrifice, which anchors building to community and place; the essential meaning of taulaga for Samoans is both anchorage and a sacrificial offering (Pratt 1893). Sacrifice, notes Mauss, ‘is a social function because sacrifice is concerned with social matters’ (Hubert and Mauss 1964: 102). This relationship between sacrifice and building is made possible via a ‘zombification’ process (Luckhurst 2015; Davis 2000; Platts 2013) in which the sacrifice of life through bloodletting reanimates architecture. With the demise of the modernist paradigm of the building as machine, the notion of reanimation has become critical to contemporary art and architectural discourse largely through two advocates of the agency of art. First, the art historian and cultural theorist, Aby Warburg (1866–1929), in his research on the afterlife of antiquity, proposed that Pathosformeln (pathos formulae) persist atavistically as obsessive archaic anthropomorphisms and rituals. Second, British anthropologist, Alfred Gell (1945–1997), in his Art and Agency (1998), considers artworks, like living persons, to act upon viewers; as living presences, they exercise agency.2

98  Ross Jenner and Albert L. Refiti



As material assemblages, buildings are reanimated through the symbolism crafted and embodied in their materiality. The work of architects and masons (Europe), tufuga-faufale (Sāmoa), papa-kuhikuhipu‘uone (Hawai‘i), or tufunga langafale (Tonga) infuses buildings with this capacity to suspend life. Comparing the Western classical architectural tradition and pre-contact Polynesian buildings, we expect to show cohesively how the idea of materializing sacrifice is carried out in architecture. The term ‘zombie’ denotes a ‘living’ in quotation marks; such ‘life’ in, say, Sāmoa or Greece is comparable in so far as the traditions, both of the fale/whare and that of the classical temple, survive into the twentieth century. This is evident in examples such as the Fale Pasifika and Fale Sāmoa in Aotearoa New Zealand, and neo-Classical architecture in Europe and the Americas. We contend that blood sacrifice relates to the work of mourning and melancholia. Mourning and melancholy are described by Freud as the incorporation and internalization of the death and loss of the loved object, intimating that melancholy is a type of ‘mourning in advance’ in which a life is sacrificed in order to treat ‘the object of libidinal investment as lost while the object is still here’ (Žižek 2011): we lose a life but gain it forever by its embodiment within the architecture. This, we propose, parallels the ‘zombification’ process in ritual architecture, where the extraction of life through blood sacrifice binds local groups together via their progenitor, and the event is encrusted and inscribed on the architecture – buildings are materiality encrusted with the dead or zombies. They become ritual attractors:3 ‘embodiments of deities’, with ‘cosmological or spiritual references’ and ‘durable to semi-durable features (posts, altars, etc.) with intensified ritual focus’ (Kahn 2008: 15). Sacred buildings as ritual attractors maintain and reanimate the dead so that the living can remain-in-place. One basic experience of what is termed ‘sacred’ is sacrificial killing. The German scholar, Walter Burkert, upon whom Hersey relies heavily, notes: ‘In the experience of killing, one perceives the sacredness of life; it is nourished and perpetuated by death. This paradox is embodied, acted out, and generalized in the ritual. Whatever is to endure and be effective must pass through a sacrifice which opens and reseals the abyss of annihilation’ (Burkert 1983: 38). Hence, he notes, building-sacrifices are widespread. ‘A house, a bridge or a dam will stay strong only if something lies slaughtered beneath it’ (Burkert 1983: 39). He continues: One of the most detailed Latin descriptions of a sacrifice depicts the erection of a border-stone. A sacrificial animal would be slaughtered in a pit and burned together with offerings of incense, fruits, honey, and wine. The stone was then placed on top of the remains while they were still hot. Thereafter, neighbors would return



Zombie Architecture  99 regularly on the anniversary of that sacrifice to repeat it. Similarly, altars and statues can be set up over a victim in the course of a ritual.

There are also references to human sacrifice in Greek folk culture in a poem concerning ‘Arta’s bridge’, where the chief builder sacrificed his wife to establish a good foundation for the bridge.4 A similar legend appears in the Romanian folk poem, ‘Meșterul Manole’, in the building of a church in the earliest Wallachian capital city.5 It must be noted, however, that recent scholarship on classical sacrifice, and sacrifice in general, contests the centrality of killing as posited by Burkert (where sacrifice is justification of violence and obfuscation of guilt), along with that of meat eating, as posed by Marcel Detienne and Jean-Pierre Vernant (where suppression of violence and justification of social and political organization is ensured) (Detienne and Vernant 1989). Recent scholarship argues that such grand theories of sacrifice can no longer hold sway. Thus, Faraone and Naiden stress the fact that the word sacrificare, making holy, ‘refers to any act by which something was put into the possession of a god’, whilst immolare, ‘to sprinkle meal’ (Faraone and Naiden 2012: 4), does not refer to slaughter any more than the Greek thuein, which refers to making smoke in the form of ‘burning an offering, whether of flesh, other food, or incense’ (Naiden 2013: 278). In these accounts, slaughter appears as one end of a spectrum ranging from animal to vegetable to mineral and from libation to smoke, the main sense being that of offering and prayer. The German Opfer, offering, oblation, sacrifice, embraces this idea, as does the idea of ‘votive offering’, or ‘ex votivo’.6 Other forms of votives include foundation deposits denoting materials (including sacrificial remains) that were intentionally buried in connection with them.7 Prayer, presence and present are also important in Naiden’s proposal that the worshiper must approach the deity in a shrine, present an offering and some request that goes with it, commonly a prayer, and present him/ herself. The worshipper, like the offering, must be kalos (beautiful), physically, thanks to a garland or white clothes, and morally. The worshipper slaughters a victim, divides it between himself and a god, and burns the god’s portion (Naiden 2013: 33). Large sculptures in Greek cities, such as the sacrificial bull that stood in the Eleusinion in Athens and reliefs of bulls being led to sacrifice on the Parthenon frieze, are representations of sacrifice, but they are also substitutes for sacrifice. The Parthenon itself is not a temple, having no altar and no priests or priestesses to officiate; rather, it is a storehouse for votives. ‘It perpetuates the thanks given by the Athenians after the battle of Marathon, and so it commemorates sacrifices’ (Naiden 2013: 123). In effect, it is a substitute for sacrifice, ‘a giant gift to Athena’ (Neer 2012: 103).

100  Ross Jenner and Albert L. Refiti



Figure 6.1. (a) The Parthenon, the Theatre of Herodes Atticus in the front, and the Propylaia (the entrance to the Acropolis) in the top left. Photo by Larry from Charlottetown, PEI, Canada, 2010. CC 2.0. (b) Youths leading a heifer to the sacrifice. Block XLIV (fig. 132–136) from the South frieze of the Parthenon, ca. 447–433 BC. CC 2.5.

Language Tropes Describing Space and Architectural Ornaments Linked to Sacrifice Classical Greece The uneaten parts of a sacrifice (bones, skin, etc.) were reconstructed and arranged in rows of tables of offering, set out on wall tops, made into pendants or hung on obelisks (meaning meat spit/skewer). Whether rearranged on an altar or eaten by the congregation, the victim’s remains were clearly too holy to be thrown away or ignored (Hersey 1989: 19). There is an echo here of proto-Indo-European sacrifice, as a ritual repetition of creation, in both the social and cosmogonic senses. In these rites the victims were killed and dismembered and their bodies carved up with the greatest care. Parts of the victims’ limbs were distributed and eaten. The different value and prestige of the various cuts of meat represented the hierarchical position of the individual or group.8 In myth, it was Prometheus who, as well as bringing fire, instituted the model of sacrifice in arranging in the correct position (euthetisas) the bones to offer to the gods (Hesiod, Theogony: 541). This corresponds with that of the Persian magus, whom Herodotus describes as offering an incantation recounting the birth of the gods, a cosmogony, after carefully disposing (diatithen) all the meat of the sacrificed animal on a bed of clover (The Histories, 1: 132). Other parts of the body, bone and fat, were dispersed into the cosmos, often by fire, to transfer material substance from microcosm to macrocosm.



Zombie Architecture  101

The dismemberment of the sacrificial victim corresponds to that of an architectural ordering of components, carefully disposed and re-membered as parts in a part/whole relationship, and proportioned, in Vitruvius’ account, anthropomorphically. However, the temple is also polysemous: ‘we can see the temple as a grove of sacred trees decorated with battle or hunting trophies, or decked out like an altar, with reconstructed sacrifices. Rows of teeth, garlands, horns, bones, weapons, and other things taken from victims figure in such displays, as do flowers, fruit, and the like’ (Hersey 1989: 21). For Hersey, ‘[t]he names for the different elements of a temple’s architectural order – the decorations of base, shaft, capital, architrave, frieze, cornice, and pediment – bear this out’ (Hersey 1989: 21). The binding of victims’ feet are matched by binding motifs at column bases. The head, or capital, in the Doric order, is formed of an echinus carved or decorated with spiny leaves, which is to say, a garland, such as is commonly worn around the head. Ionic and Corinthian capitals have head garlands, including blossoms, and also hair and horns. Triglyphs are comparable to a thighbone chopped into three. The guttae, or drops beneath, drained from the thighbones, represent the sacred fluids carefully drained into the altar, or act as memorials of the different sprinklings in sacrifice by the priest. Vitruvius’ account has the Corinthian order originating with a basket (kalathos) left at the tomb of an aristocratic maiden at Corinth. An acanthus plant grew up and around the openwork of the basket, giving it a beautiful, leafy appearance that caught the eye of the architect Kallimachos as he passed by. Inspired, Kallimachos reportedly designed a capital imitating the image. The Corinthian order was born. As Joseph Rykwert observes, however, there are five tell-tale elements here: a virgin, a death, an offering basket, an acanthus, and the notion of rebirth (Rykwert 1996: 320).

The Pacific The Samoan fale‘ula, a ceremonial building reserved for chiefs’ meetings and looked after by the taupou, the village maiden (Krämer 1994), is named after the first sacred house of the progenitor Tagaloa-a-lagi. The house’s name derives from its being painted red (‘ula), the colour of blood from sacrificed animals and humans smeared on the posts. Blood signals the biological matter that connects the present moment to the time of the progenitor and forms a line that is protected, enclosed by tapu and transferred via the blood of sacred women, the taupou,9 whose lineage was highly sought by renowned ali‘i (high chiefs). This blood connection was celebrated when the taupou was married in a defloration ceremony called fa‘amaseiau. In the ceremony, the taupou’s hymen was ruptured publicly in the faletele (oval guesthouse) or on the malae (ceremonial cleared space)

102  Ross Jenner and Albert L. Refiti



surrounded by the two families. The vaginal blood was smeared on fine mats and the faces of her entourage (aualuma), which signalled the important connection of this biological matter to the progenitor,10 later sealed with the birth of the first child (tamasā), who would carry the biological matter on to the next generation. Before she was married, the taupou resided in the faletele, carefully watched by her entourage to protect her from being sullied by everyday things, including domestic tasks. She was shielded from the sun, sometimes confined indoors until the evening, to prevent her skin from turning dark. She was also fed with special wooden tongs to prevent her from touching food and water. The word taupou (tau meaning position and pou meaning post) refers to the position that she held in the faletele, being seated underneath the main central posts of the house. The fale‘ula was the sacred house in which she dwelled and which kept her mana (sacred) blood safely intact before being married. Elsewhere in Polynesia, sacrifice is connected to land settlement rites, in which the death of a significant person signals the ‘quieting of the land’. Valeri observes that in Hawai‘i: [the] kāli‘i rite entails the king’s ‘execution’, symbolized by the spear that touches him. … [It is] by dying in the hands of the people and the god Lonomakua that the king obtains the land. This is because death neutralizes his ‘wildness’ and makes

Figure 6.2. (a) The Origin of the Corinthian Order, engraving, 1684, from Claude Perrault’s French translation of Vitruvius. Transferred from en.wikipedia to Commons by Markellos using CommonsHelper, author unknown. (b) Zeus altar in Olympia, author Davide Mauro, 2018. CC 4.0.

Zombie Architecture  103



possible his transformation and incorporation into the community that has been newly re-formed in the festival. (Valeri 1985: 225)

The sumu (lashing ornament) of the Samoan faletele celebrates the joining of two significant elements: the vertical poutū (central post) and horizontal ‘au‘au (ridgepole). As props, the vertical posts hold together and apart the world of humans and Lagi (heaven) while the horizontal ridgepole signals the spanning cosmos. The sumu which ties them both together relates to the Southern Cross ‘cluster of stars’11 made up of the diamond-shaped sumu fish. In Samoan and Tongan mythology, the sumu (fish) and the toloa (duck) were sacrificial offerings taken up to the heavens and made into a sign (Krämer 1995: 360), which now forms a place in the night sky that navigators use to position their craft and guide their journeys on the ocean (Krämer 1995: 284). Pratt suggests that sumu also means ‘to enter the body as a spear’, in reference to the shape of the wound made by a spear (Pratt 1893: 278).

Use of Architectural Forms, Such as Mounds and Altars, as Central to Rituals of Sacrifice Both Western altars and Polynesian ahu have their origins in places of sacrifice that were central to the architecture of rituals celebrating cosmic relationships. Conventional accounts of Pacific architecture privilege the tectonic, which is to say, timber building. By emphasizing mounding here (as opposed to framing, jointing and hanging), we trust that balance may be restored and the sacrificial addressed. Mounding is found at the beginnings of European architecture, and we suggest it is also at the beginnings of Polynesian architecture.

Europe German historian and architect, Gottfried Semper defined the act of building up, heaping, piling, stacking stone or earth, or ‘stereotomy’, as the first of four fundamental elements of building: ‘If we define building up a turf mound or levelling an irregularly shaped rock as stereotomic labour, then the oldest and noblest symbols of society and civilization – the hearth, or the highest expression of the same cultural idea, the altar – may serve us as a starting point’ (Semper 2004: 726). Fire and mounding, it may be noted, are joined here from the start. Semper’s discussion then moves to forming the bases of palaces and temples – as at Tyrns, Maecenae and the Athenian Acropolis. In Greece, the bōmos (altar), or trapeza (table) is generally regarded as preceding the temple, whose typology derives from the king’s house. Within the temple, the cult

104  Ross Jenner and Albert L. Refiti



statue looked east through its doors to the altar in front it. The very word altar, the place for sacrifices, derives from the Latin altare, to raise, exalt. An altar is a high place. Burkert notes regarding the altar of Zeus at Olympia: Whereas the entrance to the precinct of Pelops is in the west, the altar of Zeus was approached from the stadium, i.e., from the east. Whereas blood was poured into the sacrificial pit for Pelops, that is to say, downward, the altar of Zeus grew higher and higher. Thus, the two sacrificial recipients were united in a polar tension. (Burkert 1983: 97)

He cites Pausanias, the Greek geographer: ‘The custom is to slaughter victims in the lower part of the altar, the so-called prothysis. Then they take the thighs up to the very highest point of the altar and burn them there’ (Pausanias, 5.13: 9–10). Thus, the foot race at Olympia presupposes the bloody act of killing: ‘The end of the race, its goal, is the top of the ancient heap of ash, the place where fire must blaze and burn up the thigh-bones. The race marks the transition from blood to purifying fire, from encountering death to the joyful satisfaction of surviving as manifested in the strength of the victor’ (Burkert 1983: 97–98). The altar was completely covered so that in effect it was made up of the ash of previous sacrifices; by the second century ad, Pausanias noted that the altar was twenty-two feet high.12

The Pacific Sacred mound architecture13 and malae in Polynesia are forms that solicit and keep divinities in a place, enabling the ali‘i (or ariki or eiki – chiefs and priests), to commune with them (Pollard 2013). Heaped mats (Sāmoa) or stone seats (Tahiti) play an important role here; they keep the place still, stopping movement – quieting the land, making it holy – sacer/tapu. The mound is the elevated ground where humans ascend to meet the ancestor gods, in which sacrifice is offered, and the blessing and protection of the gods are given in return. The most notable example of this in Polynesia are the Hawaiian heiau or haiau: places of worship made from levelled stone mounds where sacrifices were offered. The term heiau, according to Valeri, is derived from hai, ‘a sacrifice, to sacrifice’ (Valeri 1985: 172). Heiau are principally dedicated to the gods Lono (the deity of agriculture/fertility) or Kū (the deity of war). The first fruits are offered to Lono, and blood sacrifice, including human victims, are offered to Kū, in order to engender fecundity, on the one hand, and to purify the land, on the other (Valeri 1985: 176). Significant heiau contain luakini (temples) in which kings consecrated human and animal sacrifices. According to Valeri, in many luakini temples



Zombie Architecture  105

where sacrifices are made, a lele (altar) is placed in the sanctum sanctorum, an area of the temple circumscribed by a platform that has either the shape of an unu (horseshoe or ‘open mouth’ since it resembles ‘an open mouth formed by a line of stones’, Valeri 1985: 177) or of the waihau (round temple, waihau here meaning ‘a round heap’ or ‘small, tight bundle’, Valeri 1985: 176). Luakini means ‘four hundred pits’; lua are ‘pits where remains of the victims were thrown’ (Valeri 1985: 237). The pits are located either inside an anu‘u (tower) erected on one end of the heiau or other places on the platform. The malae are sacred places similar to the heiau but they indicate the centre of a settlement (unlike heiau that are generally located in sacred places in the landscape, away from people and the community). The ideal form of a malae is either oval or circular and, in our view, repeats the form of the sanctum sanctorum (waihau and unu) discussed above. We believe that the malae is a sacred space that bridges the space where sacrifice takes place, on mounds located away from the settlement, and the village proper. The malae are everyday working spaces that mirror the function and ritual forms of sacred mounds. The horseshoe form (unu) is repeated in the formation of the Tongan kava ring or the alofisā ‘ava ring in the Samoan fono (council) – both are held on the malae.14 Unu denotes also the small stones used to fasten the posts of a house when erected in the ground (Andrews 1922: 604). Our contention is that unu, as an open-mouth-form, represents the ideal shape of spaces that originated in rituals of sacrifice and are now in use as the default form for community and congregation in Polynesia. The kava ceremony15 is held at the open mouth of the ring and denotes the place between tapu and noa (restricted and unrestricted); thus, it is a space that solicits the mana of the ancestor, to enclose it (the meaning of tapu) and then to siphon it out from the ring safely for the common people to use. This is the meaning of sacrifice and architecture in Polynesia: a zombie machine that functions to suspend life and death in an endless exchange that perpetuates the mana of the ancestor gods.

Ornaments in Buildings as Concretions or Reconstructions of the Dead Within Classical European Architecture In Greek sacrifice, reconstruction took the form of reconstituting the dead parts as a way of giving them a new life. As trophies consist of a warrior’s outer teguments (breastplate, helmet, spear, etc.), so the column became

106  Ross Jenner and Albert L. Refiti



a form of preserving and recording sacrifice. Similarly, the contents of an urn are a concretion of the dead person within it. According to Hersey, the triangular pediment of the temple enclosed by mouldings, the tympanum, was originally a structure of bones covered with animal skins and used as a drum: the same materials were used to reconstruct victims on altars. Hersey compares the entablature above the columns to tables of offerings outside temples, upon which ‘flowers, fruit, vegetables, and other “victims” were arranged’ (Hersey 1989: 30).

The Pacific Perimeter posts of Polynesian sacred houses were generally considered embodiments of important ancestors. Most were carved with qualities of the ancestors, as carved posts in many Māori whare tīpuna (ancestral houses) show. The experience of being inside these houses demonstrates our point regarding the zombification process, whereby architecture aids in constructing spaces that suspend the dead and living together. Sacred houses were also places of sacrifice. T.H. Hood reported in his travel diary that Fijians and other Melanesians carried out sacrifices in which, ‘with each post of a new house, [they] buried some unfortunates alive, placed in the postholes, each standing up with arms round the tree’ (Hood 1862: 40). Augustin Krämer recounted that human skulls were unearthed underneath the centre post of a chiefs’ house in Sāmoa (Krämer 1995: 259), supporting oral stories of sacrificed persons buried under central posts.16 That the erection of significant ancestor houses required the quieting of the place with human sacrifice is also suggested by Tregear’s observations on Māori ‘grand houses, such as a temple or council-hall’ (1904: 279). His account, written for ‘the settler, the anthropologist, and the tourist’ (Preface), maintains that a member of a tribe was killed at the opening of such houses, his heart consumed by a priest, and that bodies were ‘buried inside the house at the base of the end-slab (poupou-tuarongo) next the back of the building’ or ‘at the foot of the central pillar, the pou-tokomanawa’ (280). These bodies, Tregear reports, were called ‘stone’ (whatu) since they served as foundation-stones of the new edifice.17 In Sāmoa, posts are not carved but lashed in ornamental mamanu. The word manu means animal, and ma not only denotes ‘pure’, ‘clear’ or ‘bright red’, but also ‘to be all destroyed’ (Pratt 1893). Thus, ornaments are a material concretion of the dead, and lashings tie sacrificial animals to buildings – the destruction of life for purification. The sumu18 pattern at the junction between the central post and ridge beam is a binding motif, too, in the shape of a diamond (malu), which is symbolic of the Southern Cross constellation and the shape of a spear wound.



Zombie Architecture  107

Conclusion Mourning and melancholia (Freud 1917) infuse architectural strategies that symbolize the dead, in a betwixt world of the living as petrified artefacts, and suspend ancestral time as impressions on the present. Ancestors are progenitors and gods in both Western and Pacific mythologies; our connection to them guarantees that the human world remains part of a cosmological schema. In discussing Neolithic mounds in England, Joshua Pollard refers to Polynesian ahu and marae (marae and malae designate the same phenomenon) to stimulate an understanding of how their coming into being must be located within ‘flows of matter, ideas, people, and spiritual agencies’ (Pollard 2013: 190–91). Thus were the British henges, he suggests, endowed with movement, animated and reanimated. Such an understanding demands awareness of different kinds of ontological conditions that occur through buildings and beyond them. Writing after the demise of grand theories of sacrifice, Fritz Graf notes that such ‘[r]esearch has contracted upon single cultures; cross-cultural approaches are rare. But they might be rewarding’ (Graf 2012: 51). Juxtaposed images, in this way, may help to mobilize and reframe fixed understandings. We trust that an encounter between Pacific and Western concepts of sacred architecture might evoke commonalities that could re-frame understandings of both architectural traditions. We propose that the ‘re-animated dead’ connects them both. We do not propose that such a connection can be easily established, but we do believe that certain architectural elements – processions, altars, sacrifices, platforms, posts and ornamentation – provide a lively base whereby one culture may offer food, if we might use such words, for thought, to another, backwards and forwards. Dr Ross Jenner teaches architecture at the University of Auckland. He has practised in New Zealand, the UK, Finland and Switzerland, and taught at universities in Australia and the US. His PhD is from the University of Pennsylvania under Joseph Rykwert. He was a leader of Auckland’s winning Venice Prize at the Biennale di Venezia, 1991, Commissioner for the New Zealand Section of the XIX Triennale di Milano, 1996, and is an executive editor of the journal Interstices. Having exhibited and published in the US, UK, Italy, Australia and New Zealand, he is currently researching modes of materiality in architecture. Albert L. Refiti is a research leader in the field of Pacific spatial and architectural environment with extensive research and publication experience in the area, supported by his teaching and lecturing in the last fifteen years.

108  Ross Jenner and Albert L. Refiti



His current research is on Pacific concepts of space – how they are formulated and enacted – the aim of which is to find out how this understanding might play a role in rethinking the ways that Pacific people can create new modes of working and new notions of place and citizenship in the diaspora towards a Pacific cosmopolitic.

Notes   1. Joan Breton Connelly’s controversial reading, The Parthenon Enigma, goes a step further: there, sacrifice appears at the core of the Parthenon whose frieze depicts Erechtheus’ sacrifice of his daughter to save Athens from an invasion by Eumolpos and the Eleusinians (Breton Connelly 2014: 136).   2. Other scholars, for example, Papapetros (2012), Payne (2014), Hunt (2006), van Eck (2015), have followed this line of agency through affect.   3. J.J. Fox describes ritual attractors in Austronesian houses as ‘part of the structure of the house. It may be a specific post, beam, platform, niche, altar or enclosure that has a pre-eminence among the other parts of the house and, as such, represents, in a concentrated form, the house as a whole. The rituals of the houses acknowledge this attractor, generally from the moment of construction’ (Fox 2006: 1).   4. On the Bridge of Arta, see Lawson (1964: 262–77).   5. For an overview of the various accounts of the foundation sacrifices concerning this legend, see Dundes (1996). See also Diplich (1976); Payne (2014: 308–10). On the general subject of foundation sacrifice, see Tylor (1871); Speth (1894); Sartori (1897); Haddon (1898); Hughan (1911: 762); Klusermann (1919); Westermarck (1924–1926: 461–66); Hastings (1908); Carter (2003). On human sacrifice in Ancient Greece, see Bonnechere (1994).   6. See Didi-Huberman (2006).   7. See Müller-Zeis (1994), who develops a typology of building offerings, from the Geometric to the Hellenistic period; Weikart (2002), who traces the custom from Minoan to Hellenistic times; Hunt (2006), who examines foundation deposits of sacred buildings and their Near Eastern precedents; and Rostroff (2013), who examines the saucer pyres of the Athenian agora in their chthonic and funerial nature.   8. See Lincoln (1984).   9. Few women were ever chosen to become a taupou, which came with an office (bearer of tamasā or sacred children) and a title; only families and clans who carried important names (titles) connected to the oldest titles in Sāmoa were able to bestow taupou titles.   10. See Krämer (1994: 39); Koskinen (1960: 72); Tcherkézoff (2008: 40).   11. Robert Williamson reported that Werner Von Bülow pointed out that ‘Sumu and another star called Toloa refers to a tradition according to which a former



Zombie Architecture  109

was fish and later a wild duck that risen up to heaven’, see Williamson (1933: 130); see also Pratt (1893: 278); Collocott (1922: 159). 12. See Dillon (1997). 13. Known as heiau (Hawai‘i), tia (Sāmoa), sia (Tonga). 14. The alofisā ‘ava (Sāmoa) and kava (Tonga) ring or circle is a ceremonial gathering often also held inside sacred buildings as in, for instance, the Samoan faletele. 15. Collocott noted that the Fono-ki-tangata is a meeting in Tongatapu, Tonga, in which ‘the “fono” is the food served with kava, and “fono ki tangata” means to use human flesh in this way. When the kava was presented to the god a human sacrifice accompanied it’ (Collocott 1922: 232). The ‘ava ceremony in Sāmoa originated with the slaying of Pava’s son by Tagaloa (Krämer 1994: 553). 16. Albert Wendt and Tui Atua Tupua Tamasese relayed some of these to Albert L. Refiti (personal communication, 2005). 17. Tregear does not provide information about his sources, nor does he date the customs he reports on. Given his diffusionist orientation, terms like ‘temple’ or ‘council-hall’ likely denote European models, and one wonders how much his account, too, is modelled on European precedents. 18. Mamanu, like sumu, is the name of a fish.

References Andrews, Lorrin. 1922. A Dictionary of the Hawaiian Language. Honolulu: The Board of Commissioners of Public Archives on the Territory of Hawaii. Bonnechere, Pierre. 1994. Le Sacrifice Humain en Grèce Ancienne. Bibliothèque de la Faculté de Philosophie et Lettres de l’université de Liège, Kernos supplements. Liège: Presses universitaires de Liège. Breton Connelly, Joan. 2014. The Parthenon Enigma. New York: Alfred A. Knopf. Burkert, Walter. 1983. Homo Necans: The Anthropology of Ancient Greek Sacrificial Ritual and Myth (trans. Peter Bing). Berkeley: University of California Press. Carter, Jeffrey. 2003. Understanding Religious Sacrifice: A Reader. London and New York: Continuum. Collocott, Ernest Edgar Vyvyan. 1922. Tongan Astronomy and Calendar. Honolulu: Bishop Museum Press. Davis, Wade. 2000. Passage of Darkness: The Ethnobiology of the Haitian Zombie. Chapel Hill: University of North Carolina Press. Detienne, Marcel and Jean Pierre Vernant. 1989. The Cuisine of Sacrifice among the Greeks. Chicago: University of Chicago Press. Didi-Huberman, Georges. 2006. Ex Voto. Milan: Raffaello Cortina Editore. Dillon, Matthew. 1997. Pilgrims and Pilgrimage in Ancient Greece. London: Routledge. Diplich, Hans. 1976. Das Bauopfer als dichterisches Motiv in Südosteuropa. Munich: Verlag des Südostdeutschen Kulturwerkes.

110  Ross Jenner and Albert L. Refiti



Dundes, Alan (ed.). 1996. The Walled-up Wife: A Casebook. Madison: University of Wisconsin Press. Faraone, Christopher and Fred S. Naiden (eds). 2012. Greek and Roman Animal Sacrifice: Ancient Victims, Modern Observers. Cambridge and New York: Cambridge University Press. Fox, James J. 2006. Inside Austronesian Houses: Perspectives on Domestic Designs for Living. Canberra: Australia National University Press. Freud, Sigmund. 1917. The Standard Edition of the Complete Psychological Works of Sigmund Freud, Volume XIV (1914–1916): On the History of the Psycho-Analytic Movement, Papers on Metapsychology and Other Works. London: The Hogarth Press. Gell, Alfred. 1998. Art and Agency: Towards a New Anthropological Theory. Oxford: Clarendon Press. Graf, Fritz. 2012. ‘One Generation after Burkert and Girard’, in Christopher Faraone and Fred S. Naiden (eds), Greek and Roman Animal Sacrifice: Ancient Victims, Modern Observers. Cambridge and New York: Cambridge University Press, pp. 32–51. Haddon, Alfred C. 1898. The Study of Man. London: J. Murray. Hastings, James. 1908. ‘Foundation and Foundation Rites’, in Encyclopaedia of Religion and Ethics, vol. 6. Edinburgh: T. & T. Clark. Hersey, George. 1989. The Lost Meaning of Architecture: Speculations on Ornaments from Vitruvius to Venturi. Cambridge, MA and London: The MIT Press. Hood, Thomas H. 1862. Notes of a Cruise in HMS ‘Fawn’ in the Western Pacific. Edinburgh: Edmonston and Douglas. Hubert, Henri and Marcel Mauss. 1964. Sacrifice: Its Nature and Function. Chicago: University of Chicago Press. Hughan, Willam James. 1911. ‘Builders’ Rites’, in Hugh Chisholm, Encyclopædia Britannica, vol. 4, 11th edn. Cambridge and New York: Cambridge University Press. Hunt, Gloria R. 2006. ‘Foundation Rituals and the Culture of Building in Ancient Greece’, PhD dissertation. Chapel Hill: University of North Carolina. Kahn, Jennifer G. 2008. ‘Ritual House Posts, and “House Societies” in Polynesia: Modelling Inter- and Intra- Household Variability’, Rapa Nui Journal 22(1): 14–29. Kirch, Patrick Vinton. 2010. How Chiefs Became Kings: Divine Kingship and the Rise of Archaic States in Ancient Hawai‘i. Berkeley: University of California Press. Klusemann, Kurt. 1919. Das Bauopfer: Eine ethnographisch-prähistorisch-linguistische Studie. Graz-Hamburg: self-published. Koskinen, Aarne A. 1960. Ariki, The First-Born: An Analysis of a Polynesian Chieftain Title. Helsinki: Suomalainen Tiedeakatemia, Academia Scientiarum Fennica.



Zombie Architecture  111

Krämer, Augustine. 1994. The Samoa Islands: An Outline of a Monograph with Particular Consideration of German Samoa, vol. 1 (trans. T. Verhaaren). Auckland: Polynesian Press. . 1995. The Samoa Islands: An Outline of a Monograph with Particular Consideration of German Samoa, vol. 2 (trans. T. Verhaaren). Auckland: Polynesian Press. Krauss, Friedrich Salomo. 1890. Volksglaube Und Religiöser Brauch Der Südslawen: vorwiegend nach eigenen Ermittlungen. Munster: Aschendorffsche Buchhandlung. Lawson, John Cuthbert. 1964. Modern Greek Folklore and Ancient Greek Religion: A Study in Survivals. New York: University Books. Lévi-Strauss, Claude. 1966. The Savage Mind. London: Weidenfeld and Nicholson. Lincoln, Bruce. 1984. ‘Sacrificio e Creazione, Macellai e Filosofi’, in Studi Storici: Rivista Trimestrale dell’Istituto Gramsci: Rome: Istituto Gramsci, pp. 859–74. Luckhurst, Roger. 2015. Zombies: A Cultural History. London: Reaktion Books. Müller-Zeis, Rita. 1994. ‘Griechische Bauopfer und Gründungsdepots’, PhD dissertation. Saarbrücken: Universität des Saarlandes. Naiden, Fred S. 2013. Smoke Signals for the Gods: Ancient Greek Sacrifice from the Archaic Through Roman Periods. New York: Oxford University Press. Neer, Richard. 2012. ‘Sacrificing Stones: On Some Sculpture, Mostly Athenian’, in Christopher Faraone and Fred S. Naiden (eds), Greek and Roman Animal Sacrifice: Ancient Victims, Modern Observers. Cambridge and New York: Cambridge University Press, pp. 99–119. Papapetros, Spyros. 2012. On the Animation of the Inorganic: Art, Architecture, and the Extension of Life. Chicago: University of Chicago Press. Payne, Alina. 2014. ‘Living Stones, Crying Walls: The Dangers of Enlivenement in Architecture from Renaissance putti to Warburg’s Nachleben’, in Caroline Van Eck, Joris van Gastel and Elsje van Kessel (eds), The Secret Lives of Artworks: Exploring the Boundaries Between Art and Life. Leiden: Leiden University Press, pp. 308–39. Platts, Todd K. 2013. ‘Locating Zombies in the Sociology of Popular Culture’, Sociology Compass 7(7): 547–60. Pollard, Joshua. 2013. ‘From Ahu to Avebury: Monumentality, the Social, and Relational Ontologies’, in Benjamin Alberti, Andrew Meirion Jones and Joshua Pollard (eds), Archaeology After Interpretation: Returning Materials to Archaeological Theory. San Francisco: Left Coast Press, pp. 177–96. Pratt, George. 1893. A Grammar and Dictionary of the Samoan Language (3rd edn). London: The London Missionary Society. Rotroff, Susan I. 2013. Industrial Religion: The Saucer Pyres of the Athenian Agora, Hesperia Supplement 47. Athens: The American School of Classical Studies at Athens.

112  Ross Jenner and Albert L. Refiti



Rykwert, Joseph. 1996. The Dancing Column: On Order in Architecture. Cambridge, MA and London: The MIT Press. Sartori, Paul. 1897. Ueber das Bauopfer, Presented at the Meeting of the Berlin Anthropological Society, 20 November 1897. Semper, Gottfried. 2004. Style in the Technical and Tectonic Arts; or, Practical Aesthetics (trans. Harry Mallgrave and Michael Robinson). Los Angeles: Getty Research. Speth, George William. 1894. Builders Rites and Ceremonies: Two Lectures on the Folk-Lore of Masonry. Margate, UK: Keble’s Gazette. Tcherkézoff, Serge. 2008. ‘First Contacts’ in Polynesia: The Samoan Case (1722–1848), Western Misunderstandings About Sexuality and Divinity. Canberra: Australia National University Press. Tregear, Edward. 1904. The Maori Race. Wanganui: A.D. Willis. Tylor, Edward Burnett. 1871. Primitive Culture: Researches into the Development of Mythology, Philosophy, Religion, Art, and Custom, vol.1. London: J. Murray. Valeri, Valerio. 1985. Kingship and Sacrifice: Ritual and Society in Ancient Hawaii. Chicago: University of Chicago Press. Van Eck, Caroline. 2015. Art, Agency and Living Presence: From the Animated Image to the Excessive Object, vol. 16. Berlin: Walter de Gruyter GmbH & Co KG. Weikart, Stefan. 2002. ‘Griechische Bauopferrituale: Intention und Konvention. Von rituellen Handlungen im griechischen Bauwesen’, PhD dissertation. Berlin: Julius Maximilians-Universität zu Würzburg. Westermarck, Edward. 1924–1926. The Origin and Development of Moral Ideas. London: Macmillan. Williamson, Robert Wood. 1933. Religious and Cosmic Beliefs of Central Polynesia, vol. 2. Cambridge: Cambridge University Press. Žižek, Slavoj. 2011. ‘Transcript: Slavoj Žižek at St. Mark’s Bookshop’, IMPOSE magazine, The Parallax. Retrieved 20 April 2022 from https://imposemagazine.com/ music/transcript-slavoj-zizek-at-st-marks-bookshop.

7

Maunawila Heiau A Sacred Hawaiian Tempo-Spatial Structure Linking Hawai‘i and Moana Nui ♦l♦

Tēvita O. Ka‘ili

Mo‘olelo: Deep History of Moana Nui Migration

H

awaiian mo‘olelo (deep history) states that it was forty generations ago (around 1200 ad) when a powerful priest-navigator named Pa‘ao set sail from Wawau and ‘Upolu in Kahiki (islands south of Hawai‘i) and voyaged north to the Hawaiian archipelago (Kame‘eleihiwa 2016; Kirch 2012).1 Wawau and ‘Upolu are culturally significant places that share their names with other locations in Tahiti, Sāmoa, Tonga, and in other parts of Moana Nui.2 In Tahiti, the ancient name for Taha‘a is ‘Uporu and for Bora Bora is Vava‘u. ‘Upolu is an island in Sāmoa, southeast of Savai‘i, and Vava‘u is an island in Tonga, north of Tongatapu.3 Most Hawaiian scholars maintain that Pa‘ao came from ‘Uporu (Taha‘a) and Vava‘u (Bora Bora) in Tahiti (Kame‘eleihiwa 2016). However, Hawaiian scholar David Malo, King Kalākaua of Hawai‘i, and others point to ‘Upolu in Sāmoa as the point of origin for Pa‘ao (Malo 1898: 206; Kalākaua 1888: 21; Westervelt 1923: 65–78). Based on the generally accepted migration route of Moanans, from west to east, it is likely that ‘Uporu and Vava‘u in Tahiti were settled and named by voyagers from ‘Upolu and Vava‘u in Sāmoa and Tonga. This is a common practice in Moana cultures, as evident by the name Hawaiki or

114  Tēvita O. Ka‘ili



Savai‘i. The inhabitants of ‘Uporu and Vava‘u in Tahiti were most likely diasporic Sāmoans and Tongans. Thus, Pa‘ao was from Sāmoa and Tonga, or he was an expatriate Samoan or Tongan (or a descendant) living in Tahiti. Pa‘ao, according to tradition, left Upolu and Wawau because of a dispute between him and his brother Lonopele (Kamakau 1991). Aboard Pa‘ao’s ship was a powerful and renowned kāula (prophet), or taula in other Moana Nui languages, named Makuaka‘ūmana.4 Makuaka‘ūmana, according to Hawaiian elder Cy Bridges, came from a specific place known as Moaulanuiākea in Wawau and ‘Upolu, Kahiki (McGregor 2013). Makuaka‘ūmana earned a place in Pa‘ao’s wa‘a (a twin-hulled vessel) because he survived a daring jump from the high sea-cliff Ka‘akōheo to the momoa (a small projection at the stern) of Pa‘ao’s canoe (Kamakau 1991). When Pa‘ao left the shore of Kahiki, several kāula gathered at the high cliff Ka‘akōheo and called out to Pa‘ao to take them on his voyage. Pa‘ao told them that his canoe was full, and that the only space was on the momoa (Kamakau 1991). Pa‘ao challenged the kāula that if they could survive the jump from the sea-cliff Ka‘akōheo to his wa‘a, he would take them on his voyage. All of the priests who plunged from the cliff died, but only Makuaka‘ūmana survived the suicidal jump. Makuaka‘ūmana flew like a manu (bird) from the sea-cliff Ka‘akōheo and landed safely on Pa‘ao’s canoe (Poepoe 1906a; Kamakau 1991). The success of Makuaka‘ūmana’s daring sea-cliff jump was a sign of his very powerful mana (supernatural power), as a kāula.

Heiau: Sacred Megalithic Structures When they arrived in Hawai‘i, Pa‘ao remained in Kona, Hawai‘i, while Makuaka‘ūmana continued his journey to Hau‘ula, O‘ahu (Poepoe 1906b; McGregor 2013). Pa‘ao instituted a stricter form of kapu (tapu), or religious prohibition and sacredness, in Hawai‘i (Kamakau 1991). He also introduced a new architecture to Hawai‘i, which includes a rectangular-shaped heiau, a sacred place of worship (Kame‘eleihiwa 2016). Many of the remaining heiau structures in Hawai‘i today are post-Pa‘ao. In Hau‘ula, Makuaka‘ūmana became the caretaker of three megalithic heiau structures: Maunawila, Ka‘unihokahi and Kapoho (McGregor 2013).5 A heiau is a Hawaiian place of worship, temple or shrine (Tengan 2008: 240). They are shrines to gods as well as places of sanctity and for offering and prayers to akua (atua, ‘otua) and ‘aumākua (Kame‘eleihiwa 2016). Kame‘eleihiwa defines akua as ‘divine elements’ and ‘aumākua as ‘ancestral gods’ (Kame‘eleihiwa 2016). Akua are also ancestors who were deified because of their deep knowledge, skills and wisdom (Māhina 1992). Heiau structures are considered places of mana and kapu and are also the manifestations of the stratified and ranked social and political structure of Hawai‘i. They are also steeped in



A Sacred Hawaiian Tempo-Spatial Structure  115

Hawaiian mo‘olelo. Hawaiian scholar Lilikalā Kame‘eleihiwa unpacks the term heiau into two compound words: hei (to ensnare, to measure) and au (current, time). Thus, heiau, according to Kame‘eleihiwa, means to measure time and to ensnare the currents, which are akua (Kame‘eleihiwa 2016; Nu‘uhiwa 2012). Kame‘eleihiwa’s definition of heiau as a place for measuring time (and space) hints at why heiau lithic structures were celestially aligned to mark and track the movement of the sun, moon and constellations. Heiau structures come in various shapes and sizes. Some are small and others massive. Some are shaped round and square and others are rectangular. It was Pa‘ao who brought from Wawau and Upolu the rectangular shape heiau (Kame‘eleihiwa 2016). A heiau, generally, had an open courtyard, thatched houses or hale (fale, whare), upright stones (male stones), reclined stones (female stones), and carved images of the gods. The thatched houses included a hale mana for storing the temple god images, a hale waiea for housing the ‘aha cord or ‘aha hele honua (cord for binding the land and measuring the heiau), a hale umu/imu for the underground oven to bake sacred foods, and a hale pahu for storing the sacred pahu drums used in temple rituals (Thurman 2014). The hale waiea was generally located at the centre of the heiau, and it was the place for praying to the akua to dwell in the heiau (Kame‘eleihiwa 2016; Thurman 2014). The piko, navel, of the heiau was one of the most potent and sacred areas of the heiau. There was also an anu‘u or oracle tower for praying and receiving guidance from the divine (Kame‘eleihiwa 2016; Thurman 2014). The anu‘u tower generally consisted of three tiers: lana, the lowest tier for offerings; nu‘u, the second tier for high priests and attendants; mamao, the highest and most sacred tier for the high priests and paramount chiefs (Thurman 2014). In the entrance to a heiau, images of deities, known as ki‘i, stand steadfast like guardians or sentinels (see Figure 7.1). A heiau was dedicated to the deities at the entrance. There was also a stone pavement, known as kīpapa, for placing offerings, a wooden stand known as a lele, and refuse pits known as luapa‘ū (Thurman 2014). There were several types of heiau structures that were employed for diverse purposes. For example, there were heiau for human sacrifices, known as luakini, and dedicated to the akua Kū, god of war and politics (Kame‘eleihiwa 2016). Kū is known as Tū in Tahiti and Tonga, and as Tūmatauenga in Aotearoa.6 There were heiau structures for offerings to ensure good fishing; they were called ko‘a (Kame‘eleihiwa 2016). There were heiau structures for offerings to increase food crops, known as waihau, and agricultural heiau, called māpele (Kame‘eleihiwa 2016). These heiau were mostly dedicated to Lono, the god of fertility and agriculture (see Figure 7.1). Lono is known as Ro‘o in Tahiti, Rongo in Aotearoa, and

116  Tēvita O. Ka‘ili



he is probably the god Tangaloa ‘Atulongolongo in Tonga.7 There were also heiau known as Hale o Papa for female rituals and Mua for male ceremonies. Only females were permitted in a Hale o Papa, and only males could enter a Mua (Kame‘eleihiwa 2016; Tengan 2008). Lastly, there were heiau structures for treating and healing the sick and afflicted (Thurman 2014). Maunawila Heiau, according to Cy Bridges, was a heiau hō‘ola, a sacred place for treating and healing the sick, and a place of life (McGregor 2013).

Collaborative Work to Preserve Maunawila Heiau In June 2012, I was invited by Rebekah Matagi Walker, a Samoan and a cultural anthropology adjunct faculty at Brigham Young University Hawai‘i, to visit and help members of the Hau‘ula community in clearing the thick and overgrown vegetation at Maunawila Heiau. Since that initial visit, I have returned many times with Rebekah Matagi and our cultural anthropology students at Brigham Young University Hawai‘i, in collaboration with the Ko‘olauloa Hawaiian Civic Club, Hau‘ula Community Association, and Hawaiian Islands Land Trust (HILT) to permanently protect and preserve Maunawila Heiau from future development, preserve its historical features, and support Hau‘ula residents in their kuleana (responsibility) to this sacred cultural structure (see Figure 7.2). I currently serve as a member of the Maunawila Heiau Steering Committee. The Steering Committee meets

Figure 7.1. Ahu‘ena Heiau, Kona, Hawai‘i. Dedicated to Lono. © Tēvita O. Ka‘ili.



A Sacred Hawaiian Tempo-Spatial Structure  117

once a month to plan and coordinate the work at Maunawila. The work to preserve Maunawila Heiau happens on the Maunawila Conservation Day, which occurs on the second Saturday of each month. In addition to this set day, there are also other days during the week where we take community groups to Maunawila Heiau to work and provide service. Our collaborative work at Maunawila Heiau has allowed me and my students to learn about the deep history and ancient geography of Maunawila Heiau and to re-establish ancestral links between Hawai‘i and the rest of the people of Moana Nui.

Maunawila Heiau: Linking Hawai‘i to Moana Nui Based on the Indigenous Tā-Vā Philosophy of Reality (Ka‘ili, Māhina and Addo 2017), I conceptualize Maunawila Heiau as a sacred Hawaiian tā-vā, tempo-spatial, structure. In terms of tā (time), Maunawila is intricately linked to the deep history of ancient Moana Nui gods and long-distant movements in Oceanians. Moreover, in terms of vā (space), Maunawila is an integral part of the ancestral geography and structures of the ancient gods and ancestors of Moana Nui. This tā-vā, time-space arrangement, provides Maunawila Heiau (as well as most ancient sacred sites in Oceania) with the narrativity and materiality to re-connect Indigenous people from Moana Nui in contemporary tempo-spatiality. Maunawila is located in the moku, district, of Ko‘olauloa and the ahupua‘a, Hawaiian land division, of Hau‘ula in the island of O‘ahu, Hawai‘i. The district of Ko‘olauloa is significant because it is an area with places where the deities Kāne and Kanaloa lived in the past (Sterling and Summer 1978). Native Hawaiian elder, Cy Bridges, explains that Maunawila was a heiau hō‘ola, a sacred temple for healing (McGregor 2013), most likely a place for healers, such as lā‘au lapa‘au (Hawaiian herbal healers), to train and practise the art of physical and spiritual healing. Cy Bridges says that when Makuaka‘ūmana came from Kahiki, he brought sacred healing pōhaku (stones) from an ahu hō‘ola, an altar of healing, from Kahiki to Hawai‘i (McGregor 2013). Perhaps some of the pōhaku from the healing altar in Kahiki were used in the construction of Maunawila Heiau (McGregor 2013). The cultural and architectural practice of taking sacred stones from one place to another, as a way of taking the mana of one sacred place to initiate the mana of another, is a common practice in Moana Nui. There is a Ra‘iātea (Tahitian) tradition that a stone was taken from the holy Marae Vaeara‘i (Temple of Heavenly Footprint) to build new temples (Kame‘eleihiwa 2015). Marae Vaeara‘i is sacred because it is where the god Ta‘aroa (Tangaroa, Kanaloa) landed after he was born out of an egg in the cosmos. He landed and left his heavenly footprint, vaeara‘i,

118  Tēvita O. Ka‘ili



on the ground. It is the place where the first marae (temple) was erected in Havai‘i-Ra‘iātea, Tahiti, to mark the landing place of Ta‘aroa (Legend of Taaroa 2015; Kame‘eleihiwa 2015).8 The transportation of stones from Kahiki to Hawai‘i points to the constant voyaging between Hawai‘i and the rest of Moana Nui during this period.

Maunawila Heiau: A Sacred Archaeo-Astronomical Lithic Structure Makuaka‘ūmana, the caretaker of Maunawila Heiau, was not only a renowned kāula but also as a kilo, which possibly means a stargazer, astronomer or seer. The connection of Makuaka‘ūmana to astronomy is evident in the naming of a Hawaiian star after Makuaka‘ūmana (Kuokoa Home Rula 1909). It is likely that Makuaka‘ūmana, as a kilo, observed the movement of the sun, moon and stars to determine the ideal time and space for planting, healing and performing of other activities. Hawaiian astronomer, Kalei Nu‘uhiwa, explains that ancient Hawaiians observed the June solstice to mark the dry season and the political activities of the god Kū (Nu‘uhiwa 2012: 26). At the June solstice, ‘the luakini [human sacrifice] season and fishing seasons begins, the earth is green and active. For those on the wet side of the islands, vigorous planting occurs. For those on the dry side of the islands, vigorous harvesting occurs’ (Nu‘uhiwa 2012: 26). The Mākau a Maui (Fishhook of Maui [Scorpio]) constellation also appears in the night sky during this season. In contrast, Hawaiians observed the December solstice to mark the wet season and the governing activities of the god Lono (Nu‘uhiwa 2012: 25). At the December solstice, the ‘maintenance and rationing storage occurs. The makahiki season begins, the winter surf is active, and migrating animals arrive in Hawai‘i. For those on the dry side of the islands, vigorous planting occurs. For those on the wet side of the islands, mending gears occurs’ (Nu‘uhiwa 2012: 25). Wākea (Orion) constellation and Makali‘i (Pleiades) are also visible at night during this time. For the past two years, I have been working together with Native Hawaiians, local community members, and my cultural anthropology students to gather empirical astronomical data by observing the sun as it rises and sets during the solstices and equinoxes. Based on our observation, it appears that Maunawila Heiau was celestially aligned, architecturally, as a solar observatory to mark and track the movement of the sun during the solstices and equinoxes. Maunawila Heiau was probably a sacred archaeo-astronomical lithic structure. The sun, based on our observations, aligns with the centre of the entrance and the centre stone of Maunawila when it rises on the December solstice, or Ke Ala Polohiwa a Kanaloa (the Black Shining Path of Kanaloa). It also sets on the centre stone on the June



A Sacred Hawaiian Tempo-Spatial Structure  119

Figure 7.2. Maunawila Heiau is celestially aligned. Photo by University of Hawai‘i Mānoa Geography Department. Lines and labels of solstices and equinoxes added by author. The lines and Hawaiian names marking the solstices and equinoxes are based on the works of Johnson (2000: 194; 1981: 42) and Kame‘eleihiwa (2016). © Tēvita O. Ka‘ili.

120  Tēvita O. Ka‘ili



solstice, or Ke Ala Polohiwa a Kāne (the Black Shining Path of Kāne). Note that the solstices are named in Hawaiian after the paths of the deities Kāne (Tāne) and Kanaloa (Tangaroa) (Johnson 2000: 194; 1981: 42). Moreover, the eastern path of the sun as it travels from north to south and south to north is Ke Ala ‘Ula ‘a Kāne (The Bright Red Path of Kāne) (Johnson 2000: 194; 1981: 42). The western path of the sun as it moves from north to south and south to north is Ke Ala Ma‘awe‘ula a Kanaloa (The Red Track of Kanaloa) (Johnson 2000: 194; 1981: 42). Measuring the movement of the sun is essentially a sacred ritual of tracking the travel of a deity. The sun, known in Hawaiian as Kāneho‘ālani, is a manifestation of the god Kāne (Kame‘eleihiwa 2016). The alignment of the equinoxes with the east corner coral stone maybe a sign of the importance of Kāne (Tāne), Kanaloa (Tangaroa, Tangaloa), and Hina (Sina) to Maunawila Heiau. Although Kanaloa, god of the sea, is the deity that is most commonly associated with coral as his body form (kinolau), Kāne and Hina are also linked to coral in Hawaiian tradition (Beckwith 1940: 54, 219). In addition, the large number of seashells and pieces of fishing gear, such as octopus lures found at Maunawila, also point to Kanaloa as a significant deity for Maunawila Heiau. Several prominent coral slabs are found around the Maunawila Heiau complex. All of these coral lithic structures suggests that Kanaloa, Kāne and Hina as divinities of Maunawila. Coral stones are also common materials in the construction of maraes in Tahiti (Kame‘eleihiwa 2016). The sacred archaeo-astronomical Ha‘amonga ‘a Maui trilithon in Tonga was constructed from three massive coral limestones. Again, coral stones point to the ancient links and deepsea movements between Hawai‘i and other sacred places in Moana Nui. In addition to being a kāula and kilo, Makuaka‘ūmana was also a mahi‘ai, a planter, cultivator of food (Poepoe 1906b; Rice 1923: 116–17). He is known for cultivating kalo (taro), sweet potato, banana, ‘awa (kava), and kō (sugarcane). Sweet potato points to contact, whether direct or indirect, with South America, specifically Peru. Hawaiian tradition states that Makuaka‘ūmana planted sweet potato and taro for himself but cultivated ‘awa, kō and banana for his gods, Kāne and Kanaloa (Rice 1923: 116–17). Kāne and Kanaloa are known in Hawaiian tradition for finding freshwater and drinking ‘awa. In Tonga, kava and banana are associated with the Tangaloa clan. The first recorded instance of kava in Tonga is kava drinking between divine Tangaloa ‘Eitumātupu‘a and his earthly son, ‘Aho‘eitu, when they met for the first time. Makuaka‘ūmana, as indicated by his agricultural activities, was a pious worshipper and follower of the pan-Moana deities Kāne, Kanaloa, as well as Hina (Beckwith 1940: 69). It was Kāne and Kanaloa who directed Makuaka‘ūmana to worship Hina, particularly the two forms of the goddess Hina known as Hina Puku ‘Ai, Hina who



A Sacred Hawaiian Tempo-Spatial Structure  121

supplies food, and Hina Puku I‘a, Hina who supplies fishes (Rice 1923: 122). These pan-Moana deities are known as Tāne, Tangaroa/Tangaloa/ Tagaloa/Ta‘aroa, and Hina/Hine/Sina/Ina in other Moana Nui traditions (Craig 2004; Ka‘ili 2017c; Ka‘ili 2016). In Tonga, Kāne and Kanaloa are known as Tangaloa Langi and Tangaloa Mana; they are companion deities, like Kāne and Kanaloa in Hawai‘i.9 Tangaloa Langi and Tangaloa Mana also rule together over natural elements, such as rain, thunder, lightning, wind and ocean (Moala 1994: 4, 12; Ka‘ili 2017c). Hina, in Tonga, is the goddess of the moon and tapa-making (making of barkcloth by pounding the bark of the paper mulberry plant with a wooden mallet), like Hina in Hawai‘i (Ka‘ili 2016).10 In Hawaiian tradition, Kāne is the god of freshwater and sunlight, Kanaloa is the god of the ocean and all marine life, and Hina is the goddess who is associated with the moon and tapa-making (Kame‘eleihiwa 2016; Ka‘ili 2017c; Ka‘ili 2016). Kāne, Kanaloa and Hina are all associated with healing in Hawai‘i (Beckwith 1940). Thus, Maunawila Heiau, a sacred place of healing and life, was probably dedicated to Kāne, Kanaloa and Hina. With regard to Hina, there is also a female stone at Maunawila Heiau that appears to be a Pōhaku o Hina or a Stone of Hina. The stone is shaped like female genitalia, specifically the vulva of the vagina. During our visits to Maunawila, several Native Hawaiians refer to the stone as a female stone. This female stone also aligns with the rising of the sun in the June solstice. Since all our celestial observations at Maunawila occur during the daytime, we have not observed the movement of the moon (a celestial object linked to the goddess Hina) during the night. I suspect that night observation would yield alignments with the moon and several of the stars important to Moana Nui (such as Makali‘i/Pleiades) to Moana Nui. As a devout worshiper of Kāne and Kanaloa, Makuaka‘ūmana was gifted with several culturally significant items: first, the malo (male tapa loincloth) and kīhei (tapa cloth shawl) from Kāne and Kanaloa (Rice 1923: 125). Makuaka‘ūmana was probably the first person to own a malo and a kīhei in Hawai‘i; both clothing items are iconic elements of contemporary Hawaiian culture. Makuaka‘ūmana also received two supernatural items: a ‘ō‘ō, digging stick, from Kāne, and a powerful ‘auamo, carrying stick, from Kanaloa. The supernatural ‘ō‘ō and the ‘auamo assisted Makuaka‘ūmana and made it easier for him to plant crops and to carry loads of crops home. Makuaka‘ūmana was also gifted the ulua (giant trevally) fish from Kāne and Kanaloa (Beckwith 1940: 70). The ulua fish is a preferred sacrificial offering in the dedication of a new heiau (Rice 1923: 137). After residing in Hau‘ula for many years, Makuaka‘ūmana returned to Kahiki on a palaoa, a sperm whale (Poepoe 1906c; Rice 1923: 126). He was warned by his gods Kāne and Kanaloa not to go down to the beach if a whale came onshore. One day, Makuaka‘ūmana heard much commotion coming

122  Tēvita O. Ka‘ili



from the beach. A whale came onshore in Hau‘ula, and people gathered and used the back of the whale as a diving board. Makuaka‘ūmana wanted to find out the source of the commotion. So, he went down to the beach. He saw the people diving off from the back of the whale. He climbed on the back of the whale, and the whale caught him and took him back to Kahiki (Poepoe 1906c; Rice 1923: 126). The beach where Makuaka‘ūmana was taken back to Kahiki is known as Ka-Lae-o-ka-Palaoa, the Bay of the Sperm Whale (Sterling and Summers 1978: 161). Another version says that Makuaka‘ūmana was taken to Kānehunamoku, the Hidden Island of Kāne and Kanaloa, and he returned to O‘ahu (Rice 1923: 126). The story of Makuaka‘ūmana and the sperm whale is probably a metaphoric narrative for the ongoing voyaging between Hawai‘i and Kahiki, especially islands south of Hawai‘i. It points to the constant voyaging back and forth between Hawai‘i, Tahiti, Sāmoa and Tonga.

Lā‘au Ali‘i Left Hawai‘i for Sāmoa and Tonga Pa‘ao and Makuaka‘ūmana arrived in Hawai‘i at the time Lā‘au Ali‘i was the chief (Kamakau 1991). When they arrived, Lā‘au Ali‘i, according to one tradition, left Hawai‘i for Kahiki (Thrum 1923: 47; Ka‘ili 2017a). Later, Lā‘au Ali‘i’s son, Pili, returned from Kahiki, with the assistance of Pa‘ao and Makuaka‘ūmana, to become the chief of Hawai‘i (Thrum 1923: 47). Again, this story points to the ongoing movement between Hawai‘i and Kahiki. After Lā‘au Ali‘i left Hawai‘i, he disappeared from Hawaiian history but appeared in Tongan tradition as Lo‘au (Ka‘ili 2017a). Lo‘au, possibly the Hawaiian chief Lā‘au Ali‘i, is a significant figure in Tongan history. He was known as the tufunga fonua, the principal architect of the Tongan fatongia (socio-cultural obligations) and the Royal Kava Ceremony, the Taumafa Kava (Ka‘ili 2017a). He is also associated with the construction of the sacred Langi Royal Tombs and the Ha‘amonga ‘a Maui (the Burden of Maui), the coral limestone trilithon. Tongan scholars I. Futa Helu and Hūfanga ‘Ōkusitino Māhina proposed that Lo‘au probably originated from Hawai‘i or Tahiti (Helu 1986; Māhina 1992). He tarried in Sāmoa (most likely in Savai‘i) before he departed for Tonga. Hawaiian culture was Samoanized before it reached Tonga (Helu 1986; Māhina 1992). Tongan history points to Lo‘au as a foreign chief who is associated with Hawai‘i, Havai‘i (Ra‘iatea), Savai‘i (Sāmoa) and ‘Eueiki/ Havaiki (Tonga). As Pa‘ao made changes to Hawaiian culture, Lo‘au was also making profound changes to Tongan society. There are Tongan stories linking Lo‘au to Savai‘i and Havaiki (Ka‘ili 2017a). In fact, Tongan history states that the kava and tō (sugarcane) plants originated from ‘Eueiki/ Havaiki, an island in Tonga where Hawaiians camped when they visited



A Sacred Hawaiian Tempo-Spatial Structure  123

Tonga every ten years to engage in sports competitions (Ka‘ili 2017a). The late Queen Sālote Tupou III stated that Lo‘au visited and left Tonga several times (Latu 2017). Tongan archaeo-astronomer Tevita Fale argues that the Ha‘amonga ‘a Maui trilithon and the seating arrangements of the Royal Kava Ceremony mark and track the movement of the sun during the equinoxes and solstices (1990; see figure 4). Interestingly, the Ha‘amonga ‘a Maui trilithon and Maunawila Heiau are celestially aligned in the same orientation to mark the solstices and equinoxes, with the east corner marking the equinoxes and the centre marking the rising of the sun during the December solstice and the setting of the sun in the June solstice.

Trans-Indigeneity for Hawaiians, Samoans, Tongans and Other Moanans The mo‘olelo of Maunawila Heiau informs the narrativity of the materiality that reconnects old ties, weaves new connections and promotes healing among Hawaiians, Tahitians, Tongans, Samoans and other Indigenous Moana Nui people who now reside in Hawai‘i. Through the power of Moana Nui tā-vā, time-space, to link people and places in past, present and future, Maunawila Heiau inspires the fulfilling of responsibilities to sacred ancestral structures, restoring of cultural protocols, and caring for ‘āina (land). As a Tongan residing in Hawai‘i, with genealogical ties to Kāne, Kanaloa and Hina, I feel a sense of duty, fatongia, to care for ancestral places that are linked to my ancient ancestors and gods.11 To acknowledge one’s genealogy is to take on responsibility (Tengan, Ka‘ili and Fonoti 2010). Thus, as a Tongan with genealogical links to Kahiki, specifically to Vava‘u and ‘Upolu, I also have a responsibility to care for ancestral places in Moana Nui that are connected to Wawau and ‘Upolu. In Moana Nui tā-vā, the past and one’s ancestors are placed in front (mu‘a, mua) as guides and leaders for the present and future (Kame‘eleihiwa 1992; Hau‘ofa 2008; Māhina 2010; Ka‘ili 2017b). This time-space orientation was evident in the blessing of Maunawila Heiau as part of its official (re)opening in modern times. In 2014, after two years of strenuous work to clear the entangled vegetation of Maunawila, volunteers decided to conduct a blessing ceremony for Maunawila Heiau to officially open it to the public. The blessing included ‘awa, a central plant that Makuaka‘ūmana cultivated for his gods Kāne and Kanaloa. The ‘awa ceremony was unique because it harmoniously and beautifully wove together strands from kava traditions of Hawai‘i, Sāmoa and Tonga. Papali‘i Dr Failautusi Avegalio, a Samoan from Savai‘i and a holder of the Papali‘i chiefly title, organized and conducted the ‘awa ceremony. He successfully included elements from kava ceremonies from Sāmoa, Hawai‘i and Tonga. For instance, the ‘awa server was a Samoan;

124  Tēvita O. Ka‘ili



the tāno‘a (kava bowl) and the fala (pandanus mat) for the ceremony were Tongan or Fijian; the ‘awa mixer and his attendants were Hawaiians, and they wore Hawaiian kīhei, the modern version of the tapa cloth kīhei gifted to Makuaka‘ūmana by Kāne and Kanaloa. The ‘awa ceremony was beautifully executed, and it provided a way to re-establish ancient ties between Hawaiians, Samoans, Tongans and other Indigenous Moana people in Hau‘ula and the Ko‘olauloa district. It created a sense of unity among a group of Moanans that do not always get along. Samoans and Tongans felt that Maunawila Heiau was a Hawaiian place, but it had deep ties to Sāmoa, Tonga and other Moanans because of common ancestors, namely Kāne, Kanaloa, Hina and Makuaka‘ūmana. The ‘awa ceremony also marked and constituted a trans-Indigenous time and space to juxtapose and to place in talanoa (dialogue) the deep history and ancestral geography of Hawai‘i, Tahiti, Sāmoa and Tonga. Trans-Indigeneity, as an emerging concept, focuses on the strategic juxtapositioning of Indigenous traditions, knowledge, histories, arts, literature and orature to see what insights, cultural truths and possibilities might emerge. It emphasizes maintaining Indigenous particularities while reaching across Indigenous communities through travel, mobility, connections, conversations and collaborations (Allen 2012). Sacred tempo-spatial structures, such as Maunawila Heiau, with deep history and ancestral geography, created the tā (time) and vā (space) for Indigenous and diasporic Indigenous peoples to fully engage in trans-Indigeneity. Maunawila Heiau sets an example for communities with Indigenous and diasporic Indigenous members to follow and improve for present and future generations. May Kāne, Kanaloa and Hina live forever in our history and locality.

Acknowledgements I want to express my gratitude to all the many people who assisted me in my research. First, the McGregor Ohana, especially Professor Davianna McGregor and Lurline McGregor, and the Ko‘olauloa Hawaiian Civic Club, particularly aunty Ululani Bernie and Revd Dr Fran Kalama. Second, I want to thank the Hau‘ula Community Association, especially Dotty KellyPaddock, and the Maunawila Steering Community, particularly Rebekah Matagi Walker, Krista Nielsen and Rosanna Thurman. I also want to say fa‘afetai to Papali‘i Dr Failautusi Avegalio, and mahalo to Hawaiian Islands Land Trust, especially Kawika Burgess and his former employee Tina Aiu. Finally, I want to express my deepest gratitude and mahalo nunui to the Kanaka Maoli (Native Hawaiians) of Hau‘ula for sharing their Maunawila Heiau with me.



A Sacred Hawaiian Tempo-Spatial Structure  125

Tēvita O. Ka‘ili is Professor of Anthropology & Cultural Sustainability and Dean of the Faculty of Culture, Language & Performing Arts at Brigham Young University Hawai‘i.

Notes   1. Kahiki is the Hawaiian name for Tahiti and islands outside of Hawai‘i. It is also a reference to all foreign lands (Case 2015). In Tongan, Tahisi is the name for Tahiti.   2. In this chapter, I use Moana Nui as the Indigenous name for Oceania and the name Moanan for people from Oceania. Moana Nui means the Great Ocean in several Polynesian languages. The name Moana Nui highlights the oceanic culture and seascapes of the people from Oceania. Over the years, I have been inspired to use the name Moana Nui instead of Pacific (or Pasifika) by the scholarly and literary works of Hūfanga ‘Ōkusitino Māhina, the late Professor I. Futa Helu, and the late Queen Sālote Tupou III of Tonga (Ka‘ili 2017a: 23). The name Pacific (or its transliteration Pasifika) is problematic because it is a colonial imposed name and it denotes an ocean that is tranquil and peaceful. Likewise, the name Oceania is problematic because it is derived from the Greek god Okeanos (Oceanus), the great river that encircled the earth.   3. There is also a place called Vavau in ‘Upolu, Sāmoa.  4. Kāula is the Hawaiian cognate for the Tongan and Samoan word taula (priest). Taula also means an anchor of a canoe or vessel. In Tonga, the body form or avatar of a deity is called vaka, canoe/vessel. Thus, a taula appears to be the anchor for the god’s vaka, canoe.  5. Ka‘unihokahi was probably a heiau dedicated to the protector shark god Ka‘unihokahi.   6. There is also a possibility that Tū/Kū is Tangaloa Tufunga in Tonga. Tangaloa Tufunga is the god of the artisans. Hawaiian history says that Kū was also the god of the artisans (Kame‘eleihwa 2018).   7. In Tongan tradition, Tangaloa ‘Atulongolongo descended from Langi in the vaka (vessel, body form) of a Pacific Golden Plover, Kiu/Tuli‘one. He was instrumental in the creation of the first islands, ‘Eua and ‘Ata, in Tonga and the first males of Tonga.   8. The footprint of Tangaloa ‘Atulongolongo, in the form of a Pacific Golden Plover, is memorialized in the Tongan sacred and high-ranking kupesi (intricate geometric design) called Ve‘etuli – literally means the foot of the Tuli/ Pacific Golden Plover (Koya 2013).  9. Tāne, in the Ra‘iātea tradition, is the descendant of Ta‘aroa/Tangaloa (Kame‘eleihiwa 2016). Thus, it makes sense for Tongans to refer to Tāne/Kāne as Tangaloa.

126  Tēvita O. Ka‘ili



  10. Hina is also a shark goddess in Tonga.   11. They are known in Tonga as Tangaloa Langi, Tangaloa Mana and Hina.

References Allen, Chadwick. 2012. Trans-Indigenous Methodologies for Global Native Literary Studies. Minneapolis: Regents of the University of Minnesota. Beckwith, Martha Warren. 1940. Hawaiian Mythology. London: Yale University Press. Case, Emalani. 2015. ‘I Kahiki ke Ola: In Kahiki There is Life, Ancestral Memories and Migrations in the New Pacific’, PhD dissertation. Wellington: Victoria University of Wellington. Craig, Robert D. 2004. Handbook of Polynesian Mythology. Santa Barbara, CA: ABC-CLIO Inc. Fale, Tevita. 1990. Tongan Astronomy. Nuku‘alofa: Polynesia Eyes Foundation. Hau‘ofa, Epeli. 2008. We Are the Ocean: Selected Works. Honolulu: University of Hawai‘i Press. Helu, ‘Ilaisa Futa. 1986. Kings and Tombs, Tongan Culture Summaries. Tonga: ‘Atenisi University. Johnson, Rubelite Kawena Kinney. 1981. Kumulipo: Hawaiian Hymn of Creation. Honolulu: Topgallant Publishing Co. Ltd. . 2000. The Kumulipo Mind: A Global Heritage in the Polynesian Creation Myth. Honolulu: Rubellite K. Johnson. Ka‘ili, Tēvita. O. 2016. ‘Goddess Hina: The Missing Heroine from Disney’s Moana’, Huffington Post, 6 December. . 2017a. Marking Indigeneity: The Tongan Arts of Sociospatial Relations. Tucson: University of Arizona Press. . 2017b. ‘Tāvani: Intertwining Tā and Vā in Tongan Reality and Philology’, Pacific Studies 40(1/2). . 2017c. ‘Tangaloa: Reviving the Path of a Moana Deity on Winter Solstice’, Huffington Post, 27 December. Ka‘ili, Tēvita. O., ‘O. Māhina and P.-A. Addo. 2017. ‘Introduction: Tā-Vā (TimeSpace): The Birth of an Indigenous Moana Theory’, Special Issue. Tā-Vā (TimeSpace) Theory of Reality, Pacific Studies 40(1/2). Kalakaua, David. 1888. The Legends and Myths of Hawaii / The Fables and Folk-lore of a Strange People. New York: Charles Webster & Co. Kamakau, Samuel Manaiakalani. 1991. Tales and Traditions of the People of Old / Nā Mo‘olelo a ka Po‘e Kahiko. Honolulu: Bishop Museum Press. Kame‘eleihiwa, Lilikalā. 1992. Native Land and Foreign Desires: Pehea la e pono ai? Honolulu: Bishop Museum Press. . 2015. Lecture #12: Tahiti Splits from Ra‘iātea, Tāne and the Manahune Tradition of Tahiti, Comes North to O‘ahu, Kaua‘i, and Moloka‘i.



A Sacred Hawaiian Tempo-Spatial Structure  127

[Online videos] Retrieved 20 April 2022 from http://www.avakonohiki. org/341-videos.html. . 2016. Lecture #8. Temple Overview of Hawai‘i and Hawai‘inuiākea: Heiau: Sacred Spaces for Learning Ancient Ancestral Concepts. [Online videos] Retrieved 20 April 2022 from http://www.avakonohiki.org/270-videos.html. . 2018. Kū/Tū Before and After the Arrival of Pa‘ao of Ra‘iātea. [Online videos] Retrieved 20 April 2022 from http://www.avakonohiki.org/270videos.html. Kirch, Patrick Vinton. 2012. A Shark Going Inland Is My Chief: The Island Civilization of Ancient Hawai‘i. Berkeley: University of California Press. Koya, Cresantia Frances. 2013. ‘Tapa mo Tatau: An Exploration of Pacific Conceptions of ESD Through a Study of Samoan and Tongan Motifs’, PhD dissertation. Suva: The University of the South Pacific. Kuokoa Home Rula. 1909. The Hawaiian Astronomy. Kuokoa Home Rula. Ka Hiwahiwa a ka Lahui. May 28, 1909 (Hawaiian Newspaper). Latu, Paula Onoafe. 2017. ‘Ko e Tala-Tukufakaholo ‘o Tonga: An Alter-Native Holistic Historiography of Tonga History From Their Own Traditional Oral Culture and Through Their Own People’s Eyes’, PhD dissertation. Christchurch, Aotearoa: University of Canterbury. Legend of Taaroa. (2015). Legend of Taaroa (Taaroa i Taputapuatea). Iaorana Motion. [Online video] Retrieved 20 April 2022 from https://youtu. be/5eRtDAnXDnw. Māhina, ‘Okusitino. 1992. The Tongan Traditional History Tala-ē-Fonua: A Vernacular Ecology-Centred Historico-Cultural Concept. PhD dissertation. Canberra: Australian National University. . 2010. ‘Tā, Vā and Moana: Temporality, Spatiality and Indigeneity’, Pacific Studies 33(2/3): 168–202. Malo, David. 1898. Hawaiian Antiquities (Moolelo Hawaii). Translated from Hawaiian by Dr N.B. Emerson. Honolulu: Hawaiian Gazette Co. Ltd. McGregor, Davianna P. 2013. Preliminary Report on Maunawila Heiau. Unpublished report produced for the Maunawila Heiau Steering Committee (based on interview with Native Hawaiian elder Cy Bridges, by Professor McGregor Davianna, University of Hawai‘i at Mānoa, Ethnic Studies). Dated January 2013. Moala, Masiu. 1994. ‘Efinangá: Ko e Ngaahi Tala Mo e Anga Fakafonua ‘o Tongá. Kolomotu‘a, Nuku‘alofa, Tonga: Lali Publications.  Nu‘uhiwa, Kalei. 2012. Hāpaiali‘i: Tracking Time. Kamehameha Schools - Keauhou Kahalu‘u Educational Group, Kahalu‘u Mānowai. [PowerPoint Presentation]. Retrieved 24 April 2022 from https://vimeo.com/49345007. Poepoe, Joseph Moku‘ōhai. 1906a. ‘Moolelo Hawaii Kahiko’, Ka Na‘i Aupuni, 7 September [Hawaiian language newspaper].

128  Tēvita O. Ka‘ili



. 1906b. ‘Moolelo Hawaii Kahiko’, Ka Na‘i Aupuni, 13 September [Hawaiian language newspaper]. . 1906c. ‘Moolelo Hawaii Kahiko’, Ka Na‘i Aupuni, 14 September [Hawaiian language newspaper]. Rice, William Hyde. 1923. Hawaiian Legends. Honolulu: Bishop Museum. Sterling, Elspeth and Catherine C. Summer. 1978. Sites of Oahu. Honolulu: Bishop Museum. Tengan, Ty P. Kawika. 2008. Native Men Remade: Gender and Nation in Contemporary Hawai‘i. Durham, NC: Duke University Press. Tengan, Ty P. Kawika, Tēvita O. Ka‘ili and Rochelle Fonoti. 2010. ‘Genealogies: Articulating Indigenous Anthropology in/of Oceania’, Pacific Studies 33(2/3): 139–67. Thrum, Thomas G. 1932. More Hawaiian Folk Tales: A Collection of Native Legends and Traditions. Chicago: A.C. McClurg & Co. Thurman, Rosanna Mari Runyon. 2014. ‘Archaeological Investigations at Maunawila Heiau: Traditional Hawai‘i in Hau‘ula’s Backyard’, Master of Arts Thesis. Honolulu: University of Hawai‘i Mānoa. Valeri, Valerio. 1985. Kingship and Sacrifice: Ritual and Society of Ancient Hawaii. Translated by Paul Wissing. Chicago: University of Chicago Press. Westervelt, William Drake. 1923. Hawaiian Historical Legends. New York: Fleming H. Revell Co.

8

Aelon Kein Ad A Case Study of Rimajol Place Identity in the United States ♦l♦

James Miller

Introduction

V

ince Diaz and Kehaulani Kauanui introduce us to the concept of ‘roots and routes’ in the exploration of Pacific diasporas to draw attention to movement within Indigenous identities (Diaz and Kauanui 2001). Pasifika, Oceania, Moana – we have always been on the move, transporting ways of life and patterns of building within our communities, transplanting new sprouts along our routes. With the rising frequency of environmentally triggered population displacement, communities are once again on the move. To reframe the vulnerabilities brought on by climate change, with its complex environmental injustices, the notions of Aelon Kein Ad (Our Islands) and vā moana, as the relational space in our Sea of Islands, provide a focal lens that allows us to explore the inherent resilience of Indigenous communities under displacement and resettlement. Warner et al. (2009) argue that to prepare for both disaster mitigation and climate change migration, we need more case studies conducted in circumstances of migration. The development of place-specific knowledge of culturally supportive systems created through local knowledge will aid in sustaining livelihoods through migration, improving community resilience. By studying the carrying capacity of a community, the kinds and

130  James Miller



degrees of resilience of embedded knowledge may emerge. This chapter provides a study of everyday place-making amongst migratory populations of the Republic of the Marshall Islands (RMI), demonstrating the continuity of cultural identity across borders and the creation of Aelon Kein Ad (Our Islands) abroad.1 The recent movement of Rimajol (people of the Aelon Kein Ad) families into Northwest Arkansas is often referred to by Marshallese as Aelon Kaal (a new island, Carpenter 2006). Aelon Kein Ad is a significant conceptual framing for Marshallese transnational place-making, as opposed to a separation of the Republic of Marshall Islands from diasporic communities. As each community in the diaspora becomes identified as ‘a new island’, Aelon Kein Ad expands to include transnational identities. My research explores the expansive relationality of Rimajol across space, focusing on the study of core socio-spatial patterns that demonstrate cultural capacity, continuity and resilience within transnational place-making between the Marshall Islands and the United States. The term transnational place-making conveys the shifting influences of Marshallese immigrants in their host communities as arrivants,2 while also accounting for the influence of host communities and remittances on home communities. Sara Lopez (2015) and Mirjana Lozanovska et al. (2013) speak to the transnational architectural influences of diasporic communities. ‘Arrivant’ is a positionality for people forced to the Americas through colonialism, imperialism and the ongoing process of settler colonialism; arrivants are settlers but not active colonizers, presenting a different relationality within host communities. As Rimajol transport ways of life, as arrivants in new host communities, spatial relationships become expansive, spiralling outward to germinate new place-identities. Not long after the independence of the Republic of the Marshall Islands (RMI) and the secession of the Trust Territories, in 1983 a Compact of Free Association (COFA) was signed with the United States. COFA allows Marshallese citizens to travel freely to the US and Marshallese citizens have been seeking education, healthcare3 and work there over the past thirty years. A chain migration brought Marshallese citizens to Seattle, Washington; Spokane, Washington; Salem, Oregon; Costa Mesa, California; Honolulu, Hawai‘i; and Dubuque, Iowa, among many other locations. Figure 8.1 depicts the dispersion of Marshallese immigrants in the United States. Recently, the predominant pattern of Marshallese migration has transformed into a labour migration to the poultry factories of Northwest Arkansas. Northwest Arkansas has a low cost of living with good schools, healthcare and jobs. The median home price for the area in 2018 was $110,000. Within the next fifty years, these Micronesian enclaves across the United States will grow, as the migration pattern transforms



A Case Study of Rimajol Place Identity in the United States  131

once again due to climate change induced displacement. As a low-lying atoll nation, the Marshall Islands are especially vulnerable to the rising sea levels and its population may soon face the status of climate refugees. The place-making of Marshallese immigrant communities in Northwest Arkansas and Oregon can be usefully examined through the lens of a transboundary relationship between rooted communities in Aelon Kein Ad and routed communities in the United States. The primary focus is on the Marshallese community of Springdale in Northwest Arkansas, because it provides a unique opportunity to study a recent immigrant community in a geographic area very different than that of their homeland; it is the largest Marshallese community outside of the Marshall Islands, with an estimated population of 15,000 Marshallese.4 Within the broader field of migration, this study begins to look at the transnational relationships between sending and receiving communities; the primary reasons for migration (economic, political, social or environmental); and the implications of translocal networks on cultural identity and place-making. The added significance of this study is that the enclaves of Marshallese in the US will potentially foster future climate refugees forced from the Marshall Islands due to rising sea levels, adding to the significance of this population (Dasgupta et al. 2009).

Figure 8.1. Map of Marshallese communities across the United States. Tacks represent known Marshallese communities. This living map was created and maintained by Carmen Chong Gum, the former Marshallese diplomat at the Springdale, Arkansas consulate. © James Miller.

132  James Miller



Place-making demonstrates the agency of immigrants on their locality (via conflict, difference, negotiation) and the production of power, meaning and new identities; that is, ‘placemaking is a vehicle for cross-cultural learning, individual agency, and collective action’ (Hou 2013: 7). The study provides a lens for understanding the spatial manifestation of cultural transference in an ethnic enclave that has recently emigrated from their homeland. As the flow of people between sending and receiving communities changes, these families are shaping both locales through trans-local place-making (Smith and Guarnizo 1998). In several cases, entrepreneurs leverage cultural resources to create economic capital in both locales, thus shaping their environs. Rimajol agency within transnational spaces either builds upon pre-migration social status or leverages capital for increased recognition within the community. In addition to Marshallese businesses, community activists have started nonprofit enterprises that provide services to the Marshallese community, such as language classes, and act as proponents for rights provided through the Compact of Free Association and agents for social justice. Rimajol communities are reshaping and redefining not only the environs of their immigrant community, but also the environment back home. Amos Rapoport provides a framework for cultural studies of migrant populations. As cultures adapt to new knowledge and technologies from outside, ‘syncretism’ takes place (Rapoport 1983). Rapoport presents the ability of a culture to carry forward a more deeply rooted cultural process despite the adoption of outside influences. An analysis of everyday living patterns and the creation of their own sense of place by Marshallese immigrants will be valuable in understanding the future development of the Marshallese communities and their physical space in the United States. The primary research questions are: do Marshallese families immigrating to Springdale, Arkansas bring a transported landscape5 with them? If so, how does this manifest in the built environment?

Methodology Employing a multi-sited case study methodology, the research focuses on the socio-spatial relationships found within the Marshallese communities of Northwest Arkansas and Oregon in the United States and Majuro Atoll and Namdrik Atoll in the Republic of the Marshall Islands. In the United States communities, the modification of existing spaces and creation of new culturally significant spaces were particularly important in uncovering cultural patterns. The study was carried out intermittently over three years (2014 to 2017), with six weeks in Springdale, Arkansas, six months in the Marshall Islands, and frequent communication with the Oregon



A Case Study of Rimajol Place Identity in the United States  133

community throughout that time period. The primary methods for data collection were participant observation, unstructured interviews and site surveys. The place-based identity and agency in the receiving Marshallese community was documented along with the identity associated with and meaning attached to key spaces and places of the Marshallese community (including homes, businesses and places of worship). Interviews were transcribed, coded and analysed using qualitative coding software in order to uncover emergent themes. The field notes from the participant observations were also analysed for emergent themes and cross-referenced with the drawings, photographs and interview analysis. Theoretical propositions helped to develop a narrative throughout the data analysis to develop an iterative mode of explanation building. Drawing comparisons between rural and urban migrations within the Marshall Islands and migrations from the Marshall Islands to the United States established understandings of transported cultural patterns in socio-spatial organization of the built envrionment (Miller 2018).

Rimajol Socio-Spatial Patterns It is clear that foreign influence, along with the project of modernity, has greatly impacted Rimajol culture and its relationship to the land on Aelon Kein Ad. The introduction of Christianity and Victorian values in the nineteenth century, the introduction of capitalism, and the Japanese and United States occupations reshaped the Rimajol socio-spatial relationship to place. The changing dynamics in the culture-environment relationship have drastically altered the landscape since the Second World War, as the occupation by the United States led to an acute period of acculturation in Majuro and Kwajalein (Hezel 1994). As migration has increased since the signing of the Compact of Free Association in 1983, these transnational relationships have further altered the manner in which Marshallese families identify with space, and the way in which place-making occurs. Given the vast change that has overcome the Marshallese way of life and the connection between culture and its design of the built environment, there are socio-spatial patterns that continue to support core cultural elements. As Rapoport’s research (1969, 1982, 1983, 2006) would suggest, the importance of the core culture in developing culturally supportive environments, the process of ‘syncretism’ and the maintenance of core culture patterns are manifest in today’s Marshallese built environment. Table 8.1 represents some core socio-spatial patterns that are indicative of Marshallese culture, identified through field research in the Marshall Islands. In the case of modified patterns, the core meaning of the space or spatial relationship has been maintained. For example, the pattern ‘A space

134  James Miller



for chatting in the cookhouse’ may not be represented through traditional materials or architecture, but the socio-cultural behaviour of bwebwenato (storytelling) continues to take place in relationship to the preparation of food. In the United States, this takes place both in the ‘kitchen’, as well as in make-shift outdoor food preparation spaces. The concept of culture-core is a starting point for understanding the culture-environment relationship within the environmental design research discourse; however, it lacks robust theoretical concepts to interpret the significance of generative patterns developed by knowledge systems, throughout the long-evolution of culture – across deep-time. Bourdieu’s (1977) concept of habitus is helpful in developing a conceptual framing for an Indigenous habitus, throughout genealogical knowledge and an ongoing relationship to ‘Land’. ‘The Evolution of the Marshallese Vernacular House’ begins to explore these interpretations of the relationship between culture and built environment that move beyond Rapoport’s culture-core (Miller 2020).

Marshallese Place-Making in the United States Following the narrative of three families (the names have been changed to maintain confidentiality) with active networks between each region and at least one atoll in the Marshall Islands, this section explores relationships between Marshallese place identity in the United States and the Republic of the Marshall Islands, highlighting the expansive nature of Aelon Kein Ad. By understanding the cultural spatial patterns inherent in everyday life that span the three regions, we can begin to understand which patterns are most important for cultural continuity within the context of the United States, and we may also begin to understand what will happen within these enclaves as they grow due to increased migration. The identity of an immigrant community is not only shaped by their relationship to one another, but also by their relationship to the various physical settings that define and structure day to day life (Proshansky, Fabian and Kaminoff 1983: 57). Therefore, the more the settings in a receiving community reflect those of the sending community, the better supported is cultural identity. Furthermore, identifying how Aelon Kaal is conceptualized among community members helps to understand how Rimajol identities are maintained within their routes. While outward facing manifestations of Rimajol settlement in Springdale are currently apparent, the growing Marshallese communities in Oregon have less observable presence within their host communities. Three main aspects represent togetherness: enra (shared resources), social relationality and memory. Marshallese housing in Arkansas and Oregon consists mostly of rented single-family homes and apartments in multi-family complexes. A few

A Case Study of Rimajol Place Identity in the United States  135



Table 8.1. List of culturally supportive, Rimajol socio-spatial patterns of habitation. © James Miller. X = pattern present, XM = pattern present but modified, - = not identified Pattern

Namdrik

Majuro

United States

The cookhouse at the centre of the Bwij (clan).

X

X

-

A space for Bwebwenato (storytelling).

X

X

XM

Local resources used in building materials for dwelling and auxillary buildings.

X

X

Three trees (pandanus, coconut, and breadfruit) providing sustenance and defining the landscape.

X

X

XM (decorations)

A central ‘big house’ for the elders and children.

X

X

XM (multi-generational house and the church as a symbol of the ‘big house’.

A cluster of familial homes around the ‘big house’.

X

XM (adapted to western tract home)

XM (multi-generational house or residing on same street)

Cultural protocol and taboo practices provide gradients of security for sleeping rooms. These define the physical construction of space.

X

X

X

Cultural protocol moderate the movement through spaces, defining the gradient between X public and private. These define the physical construction of space.

X

XM

Gathering spaces interpreted by gender and age.

X

X

X

-

136  James Miller



families own their homes, but the majority rent. The communities tend to be close-knit, forming the characteristics of an enclave. Omimono (Marshallese handicrafts) decorate the homes of almost all Marshallese families in these communities, together with framed photographs of family and ancestors, paying respect to them. These decorations also hold memories of the material culture and resources of ancestral lands, providing reminders of roots. Shoes lie by the entry door; as some young adults remarked, ‘You don’t walk inside the house with your shoes, or the shoes are going to hit you in the face’ (Interview). Some homes contain aquariums to bring back memories of the coral reefs from home on the islands (Interview). A fresh case of frozen fish from the Marshall Islands is often waiting in the deep freezer, brought over by a relative in a cooler straight from Majuro, and cans of meat and bags of rice sit in the kitchen. The common western furnishings one would expect in a home are typically missing in Marshallese homes. One might expect this is for purely economic reasons, which in some cases it is, but in others it is a carryover from the furniture arrangement back home, where sofas are not desired and ratan furniture is preferred over upholstered furnishings due to climatic impacts on material durability. Small signs of group belonging are noticeable, for example flip flops worn at school, decals of ‘Rimajol’ on car windows, and elders out on the front stoop chatting and telling stories (bwebwenato). One other defining factor is the number of individuals residing under one roof, often around ten or more, which is typical of housing patterns in the Marshall Islands and across Aelon Kein Ad. Marshallese place-making in the United States is influenced by a highly connected mesh of transboundary relationships that connect individuals and families across communities in the United States and back to ancestral atolls in the Marshall Islands. Aelon Kein Ad represents this mesh, at least conceptually. The growth of Rimajol population within Aelon Kaal is influenced by social relationships; communities rooted in specific atolls help to determine trajectories of migration. For example, in Springdale, a high number of Namdrikese (from Namdrik Atoll) have settled; decisions to move to Sprindgale from Namdrik, via Majuro and Honolulu, are often collective ones. These routes of Rimajol place-making become directed by their roots. Similarly, these socio-spatial relationships influence housing choice, the transboundary delivery of goods, and the place attachment of identities in new geographies.

The Role of the Church in Shaping Place Deacon is one of the pastors at the Marshallese Full Gospel Church in Springdale, where he has lived for the past five years. Prior to moving to



A Case Study of Rimajol Place Identity in the United States  137

Arkansas, Deacon and his twelve children lived on Namdrik Atoll, Majuro Atoll, and briefly in Costa Mesa, California (a common path of migration from Namdrik to Arkansas). Deacon lives in Springdale with his wife, daughters and grandchildren. He has children living on Namdrik, on Majuro, in Honolulu, and in California, which represents a common dispersion of Marshallese families across disparate locations from the RMI to Arkansas. Deacon moved from Namdrik Atoll to Springdale in 2013, primarily for medical reasons, and to be near his grandchildren. His children moved to Springdale for the low cost of living and the job opportunities, just as thousands of other Marshallese settling in Northwest Arkansas. Deacon and his family rent a house in a predominantly Marshallese neighbourhood, located on Wilson Street and part of a planned-unit-development in East Springdale. Back in Namdrik, Deacon was the mayor, and a school teacher before that. As is the case for most Marshallese coming from RMI, Deacon still holds land rights on Namdrik, where his ancestral roots lie. His extended family occupies the land and maintains it; however, the daughter that resides on Namdrik lives on the property of her in-laws. Deacon sends back what he can to help out his family in the Marshalls, and they return his favour by sending coolers full of tuna and reef fish to Springdale. Deacon points out that the only distinguishable Marshallese attribute of his house is the collection of shoes lying on the front porch stoop (Figure 8.2 provides a diagram of the house layout and use). The family’s living conditions are modest, and every room except the kitchen is inhabited. Similar to housing in the urban centre of Majuro, one bedroom is for Deacon and his wife, another is for a daughter and her family, another for a second daughter and her family, and the living room provides additional quarters for family passing through and children. The home functions as both a multi-generational home, which supports the passage of knowledge from elder to youth, and a receiving home along the many routes of Rimajol migration across Aelon Kein Ad. The Church is at the centre of community life, according to Deacon. In Springdale, the Marshallese church hosts significant events such as kamems (first birthdays), erak (funerary process), jebtas (dances) and holidays. Just as an alap (landholder and eldest of the family) is responsible for managing the resources and providing sustenance for those on the wāto (family land parcel), the churches are fulfilling a similar role in Springdale. They provide enra (which means both the basket of food and the process of putting together the basket for others) for the community, which also represents a resource base. This symbolism emphasizes the communal sense of sharing and selflessness, essential to Marshallese culture. ‘It’s in

138  James Miller



the language, in the resource use. It’s in the practice, cultural practices. It’s all enmeshed in this culture; all these expressions of the culture’ (Deacon). In Springdale, the pastor could be considered the alap, reshaping societal roles in the immigrant populations, and the church then becomes a symbol of both the cookhouse (the basis for sustenance) and the ‘big house’ (the house of the alap that houses the elders and the grandchildren). Given the number of Marshallese churches in Springdale (approximately thirty-two), some community members think individuals may be starting churches in order to move up the levels of social hierarchy. Similarly, alaps who move to the United States may become pastors to maintain their social position. As the pastor’s role begins to resemble that of the alap, the church becomes central to Marshallese place-making in the United States. In the context of Marshallese place-making in the United States, churches represent reconfigurations of community governance as well as a physical manifestation of Marshallese togetherness across Aelon Kein Ad.

Figure 8.2. Illustration of Deacon’s home layout demonstrating the diverse uses of spaces within the dwelling. Dwelling size is 120 metres squared. © James Miller and Purvanig Patel.



A Case Study of Rimajol Place Identity in the United States  139

Young Families Shaping Space Jack and his family recently arrived in Springdale, having come from Djarrit, Majuro. Family and friends living in Northwest Arkansas had told Jack of the jobs at Tyson Chicken and other poultry plants in the area, and he came to Springdale because he was enticed by the job opportunities and low cost of living. Jack’s brother has been in Springdale for about four years. He likes it in Springdale because of the community and the opportunities for his children. Jack’s family left Majuro when his wife was pregnant with their youngest child. Ensuring a healthy pregnancy and safe delivery is a common reason amongst Marshallese for moving to the United States; what is seen as a temporary visit then often becomes more long-term. On Majuro, Jack lived in the downtown neighbourhood known as Rita (Djarrit), which is a predominantly residential area of downtown Majuro. Many of Jack’s family members still reside in Rita, as well as other areas of Majuro where they have ancestral lands. Due to Jack’s inexperience with family finances, his family on Majuro occasionally helps with living expenses abroad. The strength of the roots in Majuro and their abundance continues to support Jack and his family in Aelon Kaal. Jack’s apartment is located in a large multi-family development near Tonitown, Arkansas, and the Tyson factories. The population of the multi-family complex primarily consists of recent immigrants, the largest part being from the Marshall Islands and Federated States of Micronesia (both of which are under a Compact of Free Association with the United States). Rent is affordable here, and the location is convenient for work, community and church. Jack’s family keeps the house decorated with Marshallese handicrafts and pictures of their extended families. There are a few plants in the house that remind them of home on the islands. Rice bags, canned meat and noodles are popular items in the kitchen, and the smell of grease from fried food is apparent. The apartment is furnished, though when there are guests, the floor provides the primary seating, just as it would be back home on Majuro. Jack’s wife stays at home with the children, while he works twelve-hour shifts at the Tyson factory. Often during the week, Jack and his wife will join friends at the Mormon Church’s basketball courts to play pick-up basketball and volleyball, while an aunty will come to watch the children. On weekends, the family enjoys having friends and relatives over for barbeques. With large meals, the preparation of meat is conducted on the front entry to minimize clean up, and the grilling takes place at the Springdale Murphy Park, where the kids can play and Jack can cook. On one occasion, approximately sixteen adults and fourteen kids filled the small apartment. Most sat on the carpeted floor, while all of the children ran around playing

140  James Miller



games. Most of the guests were relatives but a few other Marshallese friends joined, too, as well as friends from the Mormon Church. Jack calls Springdale the ‘middle ground’. He likes it because there is a large Marshallese community, and he is close to family members and friends that also came from the Marshall Islands. He also likes it because it is affordable and there are jobs. Jack would prefer to live in Hawai‘i, but the cost of living is far too high. Like many other Marshallese in the area, he misses the ocean and fishing. Jack primarily socializes at home and at the church. Typically, economic dynamics play a significant role in the selection of locations across the network of Marshallese place-making in the United States, but proximity to family and close friends that share roots in the Marshall Islands provides a significant draw to one locale. Jack’s very close childhood friend, Roger, revealed another side of Marshallese place-making in Springdale. Roger joined the military at nineteen and was stationed in San Clemente, California. After leaving the military, he moved his family to Rogers, Arkansas (Bentonville area). Roger loves California, but in order to escape the ‘party scene’ there, he decided it was best to move to Arkansas to be able to focus on studies. He is currently working on a computer science degree at Northwest Arkansas Community College (NWACC) in Bentonville. His hope is to finish his associate’s degree there and begin his bachelor’s degree at the University of Southern California. Roger misses the attachment to the ocean, but he enjoys the community aspects present in Northwest Arkansas. He came here because of family and his best friend Jack. He also sees the educational benefits for his children in the area. Nevertheless, Roger aspires to return to the Marshall Islands after finishing school and develop a computer networking business back at home. Roger enjoys attending community events, such as performances at the Rodeo. One weekend, a protestant church group had come from Rongelap to perform at the event centre located at the Springdale Rodeo.6 The performance was similar to a Jebta (traditional Marshallese dance). Due to a Latinx event taking place at the Rodeo grand stand, the Marshallese event was put on hold because of parking issues. It was not until nearly 11:30 pm that the event began. Groups of, predominantly, women were lining up to perform an entrance sequence to a crowd of approximately three hundred. The women were wearing traditional Marshallese dresses, adorned with shell necklaces and weaved hairpieces. In addition to community events, there are two known Marshallese nightlife locations in Springdale. One nightclub is adjacent the L&K Island Market; it hosts Marshallese night on Fridays and Latinx night on Saturdays. A second nightclub called Pachanga is located in a large metal warehouse that was converted into a nightclub, bar and event space. A full-height wall divides Pachanga into two dance halls – one side is oriented toward the



A Case Study of Rimajol Place Identity in the United States  141

Pacific Islander community and the other side toward the Latinx community. The music selection, dancing and activities taking place at Pachanga closely resemble those one would expect to find at clubs like Pub on Majuro. Dance styles resemble Marshallese norms as opposed to American. Dancing rarely involves close dancing, but rather demonstrating moves or dancing in groups. It is not uncommon for a group of men to dance together. Tables and chairs are arranged around the dance area, creating the sense of an audience. The style of entertainment recreates the nightlife scene of Majuro, a sense of home through music and dance. These spaces are nodes of place-making that provide familiarity for young adults who have recently moved to Arkansas to attend college or to support elder family members. Another social space that Roger enjoys is the Jake Jebol Eo store, which is a Marshallese pool hall and market, selling handicrafts, print shirts, rice, canned meat, local fish and other goods. The store is a common meeting space for community members, while the pool hall is predominantly used by men. The combination of island grocery store, Rimajol craft store and pool hall is common across larger Marshallese communities, such as at Island Market in Salem, Oregon. These markets provide familiar spaces, resembling markets in Majuro or Ebeye. Carrying foods grown on the islands or from the ocean, they create the opportunity for families and individuals to transport themselves back to home, even if just for a moment. This reconnection to place-identity encourages impromptu community meetings to occur at one of the three Marshallese stores in Springdale. Jake Jebol Eo translates as ‘provide life for others’, depicting the attitude of togetherness: a fitting name for a store that is central to the Rimajol patterns of enra and togetherness.

Living in Two Locales Mary is a Marshallese woman from Majuro, with a daughter and Micronesian husband. Mary has an American mother and a Marshallese father whose family is from Majuro. Mary’s mother, brother and brother’s family live near Salem, Oregon, while the rest of her family lives on Majuro. On Majuro, Mary’s family owns land and holds a position of power with the associated privileges – both there and in Oregon, respect is shown to them by their communities. Raised by an American mother in an urban centre, Mary lives an identifiably American lifestyle, with a few exceptions. Island populations in the Pacific that were influenced by Japanese culture and war rations typically leave slippers and shoes at the door, rice pots are readily available, and canned meat and ramen are always stocked in the cabinets. While air conditioning is now prevalent on the islands for those that can afford the electric bill, covered outdoor space is still of great value. Outside

142  James Miller



their small, western, three-bedroom tract house, chairs are placed along the breezeway facing the ocean. This is a common location for enjoying bwebwenato (chatting and storytelling) during the evening and kava with friends over the weekend. In case it rains, a carport is available, but it is not quite as pleasant because it lacks the breeze. In Salem, similar aspects are present within and outside the house. Shoes are left at the door, rice pot is at the ready, and ramen noodles are stored in the cabinet. The big difference in everyday life between Majuro and Salem is the time spent travelling from place to place, the language and the food. What persists are the close-knit relationships of family and community. Mary’s brother works with an auto parts distributor in the area and makes a side income shipping parts from Portland to Majuro. Portland is a major port for US exports to the Republic of the Marshall Islands, including construction supplies. For example, the owner of the hardware store on Majuro has a business partner in Portland, as well as family, and he constantly travels back and forth bringing Marshallese goods to Oregon and bringing back Oregon craft beers, hardware supplies, furniture and whatever else people might ask for. It is not uncommon to see the Oregon ‘O’ driving around Majuro or the ‘Ri Majol’ decal driving around Oregon given this flow of goods. Social actors, like the hardware store owner, develop new forms of capital as they pursue transnational networks, and their entrepreneurship sustains and transforms resources, including cultural resources (Smith and Guarnizo 1998: 167).

Discussion Places link to self and group identity through control, manipulation and continuities with the past. Continuity with places in one’s past is key to an individual’s place attachment, because of the significant role memories of places play in the formation and maintenance of self and group identity (Altman and Low 1992). In transporting culture-patterns, Marshallese communities recreate familiar spaces and continue to engage in collective activities that transport families and individuals back to their homeland. Marshallese communities maintain continuity in Springdale and Oregon through food, religion, close-knit community and everyday socialization, such as maintaining or reformulating social positions in the community. Diaz and Kauanui (2001) state that ‘if our roots are strong, deep, grounded, it may be preceisely for their dynamic abilities to keep pace with the variable forces of change. And then again, many have not’ (321). As Aelon Kaal represents offshoots of strong, grounded ancestral roots located in the Marshall Islands, Aelon Kein Ad becomes dynamic, with the ability to respond to the many layers of change – economic, environmental, social, etc.



A Case Study of Rimajol Place Identity in the United States  143

Additionally, priviledged social actors are able to leverage their resources in receiving communities to gain political and economic capital (Smith and Guarnizo 1998). A reciprocal relationship across transnational boundaries is significant to maintain the health of Rimajol roots. Individuals running for office often leverage their ties to transnational communities at home and abroad to support their platform. For example, during the 2016 election cycle in the Marshall Islands, family and friends of candidates campaigned in the United States, because Marshallese citizens abroad were allowed to vote, even if they have received United States citizenship. These relationships, across the routes of Aelon Kein Ad, develop a picture of Rimajol identities that are not controlled by geopolitical boundaries – further demonstrating the significance of Aelon Kein Ad as a transboundary realm. In ‘Saffron Suburbs’, Skop (2002) demonstrates the redefining of a general public space into identifiable cultural spaces among Indian immigrants in the United States. A similar transformation takes place every weekend for the Marshallese community at Tyson Park and the Springdale Rodeo. It is evident that the Marshallese communities in Arkansas and Oregon recreate places that have important connections with their past and that they imbue public spaces with cultural meaning. While these are clearly identifiable Marshallese spaces, Marshallese in the United States tend to avoid Rimajol motifs and aesthetics outside of domestic spaces. Beyond a pile of flip-flops, branded Locals or Majuros, there is little to suggest to the passer-by that a Rimajol family resides in a housing unit. Homes, businesses and community meeting places do not present an outward aesthetic that is clearly definable as Marshallese or Micronesian. Rather, distinguishing attributes are held inward. Other cultural patterns of Marshallese communities in the US that are held inward include the make-up of housing occupants and the fluidity of parenting. Marshallese tend to live in multi-generational dwellings, inclusive of extended familial relations. It is common for ten to twenty people to live in a typical two-bedroom housing unit.7 Since rental agreements seldom allow more than five individuals to live together, many families try to hide the fact that ten or more people may be living together. Based on discussions with Marshallese residents in Arkansas, there is a desire to remain hidden behind the outward appearance of a typical duplex in order to prevent the attention of authorities and possible legal ramifications. Along with multi-generational living arrangements and receiving homes, it is a cultural norm for children to move freely across familial homes. For example, one night they stay with grandma and grandpa with all of their cousins, and the next they stay with their uncle. While this cultural norm is economically effective for low-income immigrant families, it causes problems for schools when they are trying to find an absent child. Thus,

144  James Miller



Marshallese families try to keep their cultural patterns of habitation ‘in the shadows’, hidden from authorities that may want to enforce regulations and costly penalties.

Conclusion The fact that core architectural patterns, such as the ‘cookhouse’ and the ‘big house’, continue to manifest in the new places demonstrates Marshallese resilience: even as the physical manifestations of space or material change, aspects of symbolic identity persist. These patterns can help us to understand how Marshallese transnational communities’ values must inform community-based planning, and how they can provide a useful starting point for case studies investigating how climate resettlement might take shape more broadly. In considering cultural resilience, change is inherent as an adaptive strategy to disturbances, and culture change does not necessarily mean the loss of culture, but can offer ‘a creative space where new forms of cultural understanding (and practice) are developed’ (Wesson 2013: 108). Following Wesson’s argument, investments in maintaining cultural patterns to reduce the stresses on the health, well-being and security of displaced populations, while not sustaining the pre-displacement cultures, are likely to mitigate vulnerability to greater stochastic events in the post-resettlement system. Thus, we can learn to provide mechanisms that will most likely alleviate some of the culture shock and allow elements of the culture to persist – dependent on the groups’ desires in the evolution of their cultural identity. While our understanding of the transnational networks between Marshallese sending and receiving communities is still limited, it is clear that incremental changes are taking place. This is an area for continued research and analysis, especially as Northwest Arkansas is becoming ‘another atoll’ in the semiotics of Marshallese everyday life. Within Indigenous communities like the Rimajol, the relationship to land is much deeper than that defined by Proshansky, Fabian and Kaminoff (1983), demonstrating an alternate way of knowing and being in this world (Teves et al. 2015) that is difficult, if not impossible, to interpret within Western frameworks. Reflecting back on Diaz and Kauanui’s notion of ‘routes and roots’, and rethinking what land and place-identity mean for Indigenous communities, both helps to shift away from the damage-centred view of ‘ruptures’ in the culture-environment relationship and moves toward expansive and generative definitions of migration. Similarly, Goeman (2015) and Diaz (2015) demonstrate the complexity of land relations within Indigenous peoples’ identities. Through a lens of Indigenous studies, the interpretation of Aelon Kein Ad demonstrates the expansion of Marshallese relations beyond Oceania and into the American continent.



A Case Study of Rimajol Place Identity in the United States  145

James Miller is an Assistant Professor in Comparative Indigenous Studies with a joint appointment in Canadian-American Studies, Salish Sea Studies and the Huxley College of the Environment. A Kanaka Maoli scholar, architect and urbanist, James runs a design lab, ‘Ike Honua, centring Indigenous knowledge in building resilient communities through architectural and planning frameworks. He holds a PhD in Sustainable Architecture from the University of Oregon with specializations in cultural sustainability and Indigenous design knowledge.

Notes  1. Aelon Kein Ad is the Rimajol (Marshallese) term for the Republic of the Marshall Islands. Aelon Kein Ad loosely translates as ‘our islands’. The terms ‘the Republic of the Marshall Islands’, ‘RMI’ and ‘Marshall Islands’ will be used interchangeably. Aelon Kein Ad will be representative of the collective and transnational identity of Rimajol. Similarly, the Indigenous word, Rimajol will be used interchangeably with Marshallese. Rimajol translates as people of Aelon Kein Ad.   2. ‘Arrivants’, as a term for people forced to the Americas through European colonialism, was introduced through the work of Jodi Byrd (2002) and Kamau Brathwaite (1973). Arrivants are settlers, but not direct colonizers. The term has typically been associated with Black slaves brought to the Americas by force, but has been expanded to include others forced out of circumstance through the powers of colonialism, neocolonialism and imperialism. Several settler colonial and Indigenous scholars explore the the use of the term in identifying immigrant positionalities, such as Ramirez (2007) in Native Hubs.   3. An important reason to seek health care in the USA is the consequence of sixty-seven nuclear tests conducted by the USA between 1946 and 1958 on the Bikini Atoll and near Enewetak Atoll (https://www.atomicheritage.org/ location/marshall-islands, accessed 20 April 2022).  4. See www.mei.ngo, accessed 20 April 2022.   5. Transported landscapes are the cultural landscapes that communities bring with them during resettlement. It is the landscape of familiarity that provides a level of psychological comfort (Anderson 2008).   6. It is common for dance groups and sports teams to be sponsored by churches in order to travel across Aelon Kein Ad. Teams may fly in from Majuro, Costa Mesa, California, and Everett, Washington in order to play at a tournament in Springdale, Arkansas. This tends to occur in tandem with large celebrations, most notably Flag Day.   7. Ancedotally, I have heard stories of as many as twenty Marshallese family members residing in a single-room dwelling in Kalihi Valley.

146  James Miller



References Altman, Irwin and Setha Low (eds). 1992. Place Attachment. New York: Plenum Press. Anderson, Atholl. 2008. ‘The Rat and the Octopus: Initial Human Colonization and the Prehistoric Introduction of Domestic Animals to Remote Oceania’, Biological Invasions 11(7): 1503–19. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. Brathwaite, Kamau. 1973. The Arrivants: A New World Trilogy. Oxford: Oxford University Press. Byrd, Jodi. 2002. ‘Colonialism’s Cacophony: Natives and Arrivants at the Limits of Postcolonial Theory’, PhD dissertation. Iowa City: The University of Iowa. Carpenter, Dale. 2006. ‘A New Island: The Marshallese in Arkansas’, Educational Television Network. Dasgupta, Susmita, Benoit Laplante, Siobhan Murray and David Wheeler. 2009. ‘Sea-Level Rise and Storm Surges: A Comparative Analysis of Impacts in Developing Countries’, World Bank Policy Research Working Paper Series. Washington, DC: World Bank. Diaz, Vincente M. 2015. ‘No Island Is an Island’, in S.N. Teves, A. Smith and M.H. Raheja (eds), Native Studies Keywords. Tucson: University of Arizona Press, pp. 90–108. Diaz, Vincente M. and J. Kēhaulani Kauanui. 2001. ‘Native Pacific Cultural Studies on the Edge’, The Contemporary Pacific 13(2): 315–42. Hezel, Francis X. 1994. The First Taint of Civilization: A History of the Caroline and Marshall Islands in Pre-Colonial Days, 1521–1885. Honolulu: University of Hawaii Press. Hou, Jeffrey (ed.). 2013. Transcultural Cities: Border Crossing and Placemaking. New York: Routledge. Goeman, Mishuana. 2015. ‘Land as Life: Unsettling the Logics of Containment’, in Stephanie Nohelani Teves, Andrea Smith and Michelle Raheja (eds), Native Studies Keywords. Tucson: University of Arizona Press, pp. 71–89. Lopez, Sarah Lynn. 2015. The Remittance Landscape: Spaces of Migration in Rural Mexico and Urban USA. Chicago: University of Chicago Press. Lozanovska, Marta, Iris Levin and Maria Victoria Gantala. 2013. ‘Is the Migrant House in Australia an Australian Vernacular Architecture?’, Traditional Dwellings and Settlements Review 24(2): 65–78. Miller, James. 2018. ‘The Continuity of Deep-Cultural Patterns: A Case Study of Three Marshallese Communities’, PhD dissertation. Eugene: University of Oregon. . 2020. ‘The Evolution of the Marshallese Vernacular House’, Fabrications 30(1): 110–36.



A Case Study of Rimajol Place Identity in the United States  147

Proshansky, Harold M., Abbe K. Fabian and Robert Kaminoff. 1983. ‘PlaceIdentity: Physical World Socialization of the Self ’, Journal of Environmental Psychology 3(1): 57–83. Ramirez, Renya. 2007. Native Hubs. Durham, NC: Duke University Press. Rapoport, Amos. 1969. House Form and Culture: Foundations of Cultural Geography. Englewood Cliffs, NJ: Prentice-Hall Inc. . 1982. The Meaning of the Built Environment: A Nonverbal Communication Approach. Beverly Hills: Sage Publications. . 1983. ‘Development, Culture Change and Supportive Design’, Habitat International 7(5–6): 249–68. . 2006. ‘Traditional Environments, Culture and Preservation’, Open House International 31(4): 6–14. Skop, Emily H. 2002. ‘Saffron Suburbs: Indian Immigrant Community Formation in Phoenix’, PhD dissertation. Tuscon: Arizona State University. Smith, Michael P. and Luis Eduardo Guarnizo (eds). 1998. Transnationalism from Below: Comparative Urban and Community Research. New Brunswick: Transaction Publishers. Teves, Stephanie Nohelani, Andrea Smith and Michelle Raheja (eds). 2015. ‘Introduction’, in In Native Studies Keywords. Tuscon: University of Arizona Press, pp. 59–70. Warner, Koko, Mohamed Hamza, Anthony Oliver-Smith, Fabrice G. Renaud and Alex Julca. 2009. ‘Climate Change, Environmental Degradation and Migration’, Natural Hazards 55(3): 689–715. Wesson, Cameron B. 2013. ‘Rumors of Our Demise Have Been Greatly Exaggerated: Archaeological Perspectives on Culture’, Sustainability 5(1): 100–22.

9

Hinemihi 2.0 Whare-for-Export ♦l♦

Anthony Hoete

T

his chapter presents an alternative, optimistic speculation regarding the voyaging whare whakairo (carved meeting house) Hinemihi o Te Ao Tāwhito (Hinemihi of the Old World),1 ultimately re-siting her as a transformative architectural kō-design proposition rather than a static artefact. I propose that the anticipation of a possible exchange of her carvings, in return for a ‘whare-for-export’, Hinemihi o Te Ao Āpōpō (Hinemihi of the Future World) let’s say, can transform how we consider, design and create our ‘future ancients’ (Lesson 2013). The notion of a whare whakairo itself is, after all, not an ancient one, but a concept of architecture that emerged out the social upheaval relating to land loss, politics and religion during early contact between Māori and Pākeha. Since then, the whare whakairo has embodied Māori identity, manifest as built politics and optimistically imbued with the hopes and aspirations of what colonization might also have brought about. Since the signing of Te Tiriti o Waitangi (Treaty of Waitangi) by about 540 Māori rangatira (leaders) and Governor William Hobson for the British Crown (‘Signing the Treaty’ 2016), nearly 200 years ago, different kinds of relationships with Te Tiriti have profoundly influenced the fortunes of the two parties to the agreement. Perpetual breaches of Te Tiriti by settler governments have pivoted the relationships between the partners towards



Hinemihi 2.0: Whare-for-Export  149

settler interests, displacing Māori sovereignty ‘in favour of colonial hegemony, entrenching longstanding preventable inequities in health and other important domains of social life’ (Barnes and McCreanor 2019: 19). Aotearoa New Zealand has become a polarized society, marked by increasingly complex and wide-ranging systemic disparities. The Treaty (applied in an English version the rangatira did not sign) did not provide protection of Māori land and sovereignty (as guaranteed in the Māori version, which they did sign) but in fact facilitated its fast and, in many cases, forced acquisition by settlers.2 This history of turmoil raises questions such as, how might such a legacy influence the design of a whare whakairo today; who would make the decisions and how? Given such issues, a shift from individual decision-making processes towards a more collaborative wānanga approach is required.3 The importance of whakapapa (lineage) and whenua (land) undoubtedly remains paramount, but how can concepts of tikanga and whakawhanaungatanga (inclusive relations) be reconfigured to meet current needs (Ellis 2016)? How can traditions and contemporary values combine to allow creativity to flourish? For Māori, a whare (house) is more than mere shelter. It has a living presence beyond the metaphorical associations one might apply to, say, a European building. Neither effigy nor statue, an ancestral whare is not like an ancestor. It is the ancestor. Living heritage obliges a Māori orator speaking on a marae (Māori meeting place) to address and extend greetings to the house before greeting the people. From a Western perspective, talking to a building might be considered madness, like Doctor Doolittle talking to the animals (Linzey 1991). Yet, it is not appropriate to apply even well-intended Western heritage practices, including due diligence, to Indigenous cultures. Māori architecture has its own kaupapa (values), and Hinemihi can teach us to understand architecture from a South Pacific way of being in the world, rather than from a singular colonizing European perspective. A whare has a totalizing character: the self-contained spatial embodiment of genealogy, with woven, carved and painted artistic expression, carries a legacy of embracing both European and Māori forms, as well as the transformation of its community (‘Customising your Classic Wharenui’ 2008). Traditionally, an individual of significance, someone with mana (authority) and mātauranga (knowledge), would assign the ancestor to a house (D. Bonica 2021, personal communication, 11 October), often guided by a kaitiaki (guardian) of whakapapa, such as an ariki (paramount chief). In 1880, a rangatira of Ngāti Hinemihi in Te Wairoa, Āporo Wharekaniwha, decided he wanted a slice of the burgeoning Victorian tourist market at that time. Te Wairoa, with its Pink and White Terraces, was a significant tourist attraction. Āporo commissioned two carvers, Wero Taroi and Tene

150  Anthony Hoete



Waitere, to create a whare whakairo that was also to serve as a whare tapere, a place for entertaining through song and dance (which he also opened to tourists). Few meeting houses in the Te Arawa region bear female names, but Āporo chose to name the whare after the noted female ancestor, Hinemihi – famous in legend for keeping the company of a giant lizard. Wero and Tene’s carvings also represent ancestors from their whakapapa, and by including them in the meeting house, they provided a place where their spirits could dwell and protect their descendants. Both the naming of the whare and the subsequent formulation of the carvings show how ancient traditions often comprised individual actions. After the eruption of Mt Tarawera in June 1886, the entire destruction of the Pink and White Terraces, and the death of over one hundred people, Ngāti Hinemihi had to resettle elsewhere, leaving Hinemihi o te Ao Tāwhito behind. At the end of his three-year tenure as Governor of New Zealand in 1891, William Hillier Onslow purchased Hinemihi and, in January 1892, Hinemihi was dismantled, transported to Auckland by rail, and then by sea to England. Since about 1897, Hinemihi has been located in Clandon Park, facing the rear garden façade of the Onslow residence, Clandon House. Over time, she has often been visited by her communities of origin from Aotearoa, some of whose members live in London today. Thus, for some decades now, Hinemihi has connected two types of community: her Māori core community and her secondary Pākehā (non-Māori) community. These communities operate at three geographic scales: local, national and international. Regarding her Māori community, Hinemihi’s spiritual ancestry lies across Ngāti Hinemihi and Tūhourangi Tribal Authority in Aotearoa (New Zealand). In the UK, Hinemihi connects to the Māori diaspora including, but not exclusively limited to, Ngāti Rānana, Kōhanga Reo (Māori language preschool; literally, the language nest) and Te Maru o Hinemihi, a pūkenga (expert) charity, which I chaired from 2012 to 2021. In the wake of a fire in 2015, which gutted Clandon House, Hinemihi’s deteriorating carvings were taken down and put in storage. The trigger for this action was, however, not the fire, but a Listed Building Consent (Listed Building Consent 16/P/01978 2016). The Consent records my comments on behalf of Te Maru: that all the carvings must come down together (kotahitanga), that the restorative works occur transparently in the light of the day (whakatīahotanga), and that the works be done in good time (kaunukutanga). The removal of the carvings was strategic, as works carried out on them could now only be undertaken in consultation with the source community. Given the three-year timeframe imposed in the Consent, its legislative mechanism could unlock institutional procrastination, facilitate decision making and, finally, ‘see something happening’ regarding the future of Hinemihi.



Hinemihi 2.0: Whare-for-Export  151

Figure 9.1. WHAT_architecture depiction of Hinemihi envisaged without ornament. © Anthony Hoete and WHAT_architecture.

152  Anthony Hoete



At that point, the totara carvings of Hinemihi were over 140 years old. To restore Hinemihi at Clandon Park to her original dimensions, replacement carvings would be required as the existing carvings were drastically altered or have deteriorated. In 2017, as Chair of Te Maru I drafted a three-point pānui (announcement) proposing to: 1. develop a sustainable, reciprocal partnership between the National Trust (the legal owners of Hinemihi) and Ngāti Hinemihi of Tūhourangi (her spiritual owners) based on the exchange of material and knowledge; 2. connect Hinemihi to her two places (Clandon Park and Te Wairoa) and her two timeframes as Hinemihi o te Ao Tawhito and Hinemihi o te Ao Hou; 3. restore Hinemihi o te Ao Hou, or Hinemihi v2.0, at Clandon Park to her original size (that is, 50% longer and 25% taller). This process could inform the creation of the replacement carvings with their unique mauri (life principle), as well as the balance and exchange of the twenty-three existing for twenty-three new carvings. The new carvings do not all need to originate from Aotearoa New Zealand but could involve contributions from the UK. Te Maru also raised several questions: what will the bespoke carvings look like? Will they be made exclusively from New Zealand native timbers, or can we imagine British hardwoods, like oak, being utilized? Who would carve them – Hinemihi’s descendants only, or could the project be inclusive of esteemed institutions, such as the University College of London’s Institute of Archaeology, who might rigorously take on the knowledge of Māori carving, thus allowing others to participate in an Indigenous practice? In principle, the process of carving and heritage conservation could then be interpretative, inclusive, innovative and transformative. The pānui initiated a series of high-level meetings over the next two years between the National Trust, Heritage New Zealand Pouhere Taonga (NZHPT), the UK Arts Council, Historic England and Guildford Council to discuss the proposal’s feasibility. These discussions were buoyed by events elsewhere: in 2018, the Royal Academy hosted Britain’s first-ever major survey of Oceanic Art, celebrating the art of Melanesia, Micronesia and Polynesia, and encompassing the vast Pacific region from New Guinea to Easter Island, Hawai‘i to Aotearoa New Zealand. In the same year, French President, Emmanuel Macron called for the restitution of African artworks and artefacts in France: ‘I cannot accept that a large part of the cultural patrimony of several African countries is in France’ (Quinn 2018). And, of course, #BlackLivesMatter increased in popularity during 2019. In



Hinemihi 2.0: Whare-for-Export  153

the same year, the UK National Trust issued a press statement declaring a principal agreement ‘to progress an exchange whereby the historic carvings would return to New Zealand, in return for special new carvings for a Maori meeting house presence at Clandon Park’ (‘Statement regarding Hinemihi at Clandon Park in Surrey’ 2019). The creation of a whare specifically for ‘export’ is unheralded. There are four whare currently located outside Aotearoa but none were explicitly created to stand abroad. Rauru, now at the Museum für Völkerkunde in Hamburg,4 was constructed in Whakarewarewa and commissioned by a Pākeha hotelier (Tapsell 2014); Te Wharepuni a Mauī, now at the Linden Museum in Stuttgart, was originally created to represent an idealized Māori past in Christchurch (Diamond 2010); Ruatepūpuke, now at the Field Museum in Chicago (‘Maori Meeting House’ n.d.), hails from Tokomaru Bay; and Hinemihi, today at Clandon Park but previously of Te Wairoa. All four whare were created in and for Aotearoa but later shopped and shipped abroad. To produce Hinemihi 2.0 will require a process of co-design, an emerging design method for meaningful participatory collaboration. With its antecedents in the 1960s, co-design arose out of a growing global demand for more significant consideration of community concerns in decision making. In the last decades, more and more communities have concluded that architectural professionals and bureaucrats were not designing ‘for’ them but that they were designed ‘with’. In the case of the whare-for-export proposal, kō-design5 represents not just collaborative European/Māori (UK National Trust/NZHPT) research partnering. Nor can the process be led by kaupapa Māori research principles exclusively – even though there will be substantial collaboration between Māori – with and across hapū (extended kinship group) and iwi (group of hapū). The collaborations need to be articulated to align with three scales of stakeholders that correspond to the twin legacies of Hinemihi: locally, both in the UK (Ngāti Rānana, Guildford Borough Council, UCL) and in Aotearoa (Ngāti Hinemihi, Ngā Kohinga Whakairo o Hinemihi); nationally, in the UK (National Trust, Historic England) and in Aotearoa (NZHPT); and globally. These collaborations project Te Ao Māori as transnational Indigeneity and as a critical component of global heritage. In various locations and at different levels, an inclusive participatory process could ask stakeholders across all three scales to participate in the interpretation and narrative formation of the new carvings and to create a genuinely pan-iwi kō-designed whare. Hinemihi’s story already participates in a shared narrative involving multiple voices and influences. In 1979, for example, repairs to Hinemihi were undertaken by British experts who specialized in restoring historic wooden buildings. The restorers had little visual material to refer to, apart

154  Anthony Hoete



from old Burton Brothers photography taken a few days after the eruption, when her roof was covered in volcanic debris. Mistaking several tons of rooftop ash for traditional English thatch, the restorers replaced the roof with a thick covering of Norfolk reeds. This replacement need not be seen as an ‘error’. In Māori contexts, the craft can become crafty and assume material intelligence: thatched rooves are created from local natural materials, much like Hinemihi’s original roof. Although heritage organizations typically prioritize the ‘original’, there is a counter-argument regarding features arising from accumulated history. Given that Hinemihi has today spent more time in the UK (127 years) than in Aotearoa New Zealand (8 years), the inheritance of a thatched roof can be valued as a more authentic korowai (cloak) for Hinemihi than the ‘original’ shingles – which were, at the time, new building material in Aotearoa. While purists might consider the thatched roof a dilution of the cultural object, living heritage principles stress the certainty of change. In Māori architecture, it is often said that a whare is ‘carved’, not built. Taking this as a metaphor to think with, one can ask what kind of reputation could be carved out for the whare-for-export? What design questions and evolving tikanga might frame and underpin the writing of a shared narrative, as the whakapapa of the house? • What is the whakapapa of new whare? • Would both whare be called Hinemihi? • Whakapapa customarily establishes hierarchical relationships – what might this be between these particular whare? • Where will Hinemihi’s mauri reside: at Clandon Park, at Te Wairoa, or in both places? Or would the mauri lie across the broader South Pacific, if the aspiration of a whare-for-export is to whakapapa to a Londonbased Polynesian diaspora? • What would the relationship be to the two other whare called Hinemihi?6 • Given the different types of whakapapa, the technique of tararere (tracing a single line of descent from an ancestor) abandons lateral links to spouses, asking instead: does there need to be a father? (Te Rito 2007) • What is the UK National Trust’s vision for Hinemihi at Clandon? • What is the mātauranga relationship between ornamentation and structure? • What contribution could the UK make to carving production? • Could Māori carvers teach carving in the UK? Is this form of imparting knowledge permissible? • As the new carvings have their own mauri, what signs of Britishness could be included, if any?



Hinemihi 2.0: Whare-for-Export  155

Figure 9.2. Marae-as-whare: proposal for Hinemihi 2.0 integrating wharepuni (sleeping), wharekai (eating, dining) and wharepaku (toilet). © Anthony Hoete and WHAT_architecture.

156  Anthony Hoete



• Is oak an appropriate material? • What about non-binary wharepaku? • Given that digital platforms could, for example, afford multiple and simultaneous remote forms of engagement, are there innovative technologies that could be embraced? • Due to pressures to preserve the picturesque heritage of Clandon Park, one innovative proposal (WHAT_architecture Design Access Statement 2017), endorsed by NZHPT, configured Hinemihi the whare as part of a marae, integrating wharenui (meeting house), wharepaku and wharekai under the same roof – a constellation that contradicts customary building practices. How much may a contextual shift allow custom to adapt for the sake of the survival of a tāonga (culturally valuable artefact or idea)? • Who can start to write this narrative, and how, which will then inform Hinemihi’s future heritage? Trans-cultural and participatory approaches to kō-designing a wharefor-export will require significant adjustments to the traditional roles within a Te Arawa context. Nevertheless, it is possible to maintain tikanga (custom as the right way of doing the right thing for the people, at a given time).7 Crucially, the notion of a singular source community or hapū authority will need to be eschewed in favour of a more inclusive process. These lines of enquiry are intended to show that the cultural values assigned to the adorned wharenui need not be considered by Māori or Pākeha as static, in their conception or delivery, but, rather, that they can shift with the times. The whare whakairo was an ad hoc response to colonization – but as culture in Aotearoa is transcolonizing (Lukaszyk n.d.), so can architectural tikanga transcend the formulization of contemporary whare. The whare of the future thus can demonstrate how tikanga is open to interpretation, increasingly inclusive, innovative in ways not originally anticipated, and, ultimately as transformative as the culture which it aspires to represent. Whilst her whakairo (carvings) constitute less than 5 per cent of her materiality, they constitute more than 95 per cent of her Māori identity. Without them, Hinemihi is merely a grass hut. Anthony Hoete is a Professor of Architecture (Māori) and is of Ngāti Awa and Ngāti Ranana descent. He has practised in Aotearoa New Zealand, Italy, Germany and the Netherlands. In 2002 he founded his own practice, WHAT_architecture, in London, which has received Royal Institute of British Architects, New Zealand Institute of Architects and many other architectural awards for its innovative projects.



Hinemihi 2.0: Whare-for-Export  157

Notes   1. See Engels-Schwarzpaul (Chapter 10 in this volume) for more information about Hinemihi’s history and context.   2. While Māori still controlled 80% of the land in the North Island in 1860, this proportion was diminished to 40% in 1890, 27% in 1910, and 9% in 1939. The South Island land loss occurred much faster: by 1865, 99% of the South Island had been alienated. https://nzhistory.govt.nz/media/interactive/ maori-land-1860-2000, retrieved 20 April 2022.  3. A wānanga is characterized by teaching and research that maintains, advances and disseminates knowledge and develops intellectual independence, and assists the application of knowledge regarding ahuatanga Māori (Māori tradition) according to tikanga Māori (Māori custom).   4. For more detail, see Engels-Schwarzpaul in this volume (Chapter 10).  5. Kō-design is a term coined by the author to denote Māori collaborative design: from both kōrero (to talk) and kō (to locate, to dig).  6. There are three whare named Hinemihi after the eponymous ancestor of Ngāti Hinemihi: Hinemihi te Ao Tāwhito at Wairoa; Hinemihi ki Whakarewrewa (or Hinemihi II), who was carved by Tene Waitere in 1928 for his granddaughter Guide Rangi; and Hinemihi of Ngāpuna (also known as Hinemihi ki Te Paparere-a-Rātōrua), carved in the 1960s.   7. Valance Smith, in an online talanoa with the Vā Moana cluster, Auckland, 11 June 2020.

References Barnes, Helen Moewaka and Tim McCreanor. 2019. ‘Colonisation, Hauora and Whenua in Aotearoa’, Journal of The Royal Society of New Zealand 49(1): 19–33. ‘Customising your Classic Wharenui: Conservation and Adaptation of Marae Heritage Buildings’, Te Paerangi National Seminar, 4–5 March 2008. Wellington, Museum of New Zealand Te Papa Tongarewa. Retrieved 8 November 2021 from https://www.tepapa.govt.nz/sites/default/files/deanwhitingpresentation.pdf. Diamond, Paul. 2010. ‘Te tāpoi Māori – Māori Tourism – 20th-Century Māori Tourism’, Te Ara – the Encyclopedia of New Zealand. Retrieved 8 November 2021 from http://www.TeAra.govt.nz/en/photograph/23716/te-wharepuni-a-maui. Ellis, Ngarino. 2016. A Whakapapa of Tradition: One Hundred Years of Ngati Porou Carving, 1830–1930. Auckland: Auckland University Press. Lesson, Luka. 2013. ‘The Future Ancients’. Youtube. Retrieved 29 November 2021 from https://www.youtube.com/watch?v=cSENXIZYj_E.

158  Anthony Hoete



Linzey, Michael. 1991. ‘Speaking To and Talking About: Maori Architecture’, Interstices: Journal of Architecture and Related Arts 1: 50–61. Lukaszyk, Ewa A. n.d. ‘We are all Natives: Transindigenous Studies and the Cause of Global Solidarity’. Retrieved 8 November 2021 from https://www.ewa-lukaszyk.com/. ‘Maori Meeting House, Ruatepupuke II’. n.d. Field Museum. Retrieved 8 November 2021 from https://www.fieldmuseum.org/exhibitions/ maori-meeting-house-ruatepupuke-ii. Quinn, Annalisa. 2018. ‘After a Promise to Return African Artifacts, France Moves Toward a Plan’, The New York Times. Retrieved 8 November 2021 from https:// www.nytimes.com/2018/03/06/arts/design/france-restitution-african-artifacts.html. ‘Signing the treaty’. 2016. New Zealand History, Ministry for Culture and Heritage. Retrieved 8 November 2021 from https://nzhistory.govt.nz/politics/treaty/ making-the-treaty/signing-the-treaty. ‘Statement regarding Hinemihi at Clandon Park in Surrey’. 2019. National Trust. Retrieved 8 November 2021 from https://www.nationaltrust.org.uk/ press-release/statement-regarding-hinemihi-at-clandon-park-in-surrey. Tapsell, Paul. 2014. ‘Māori and Museums – Ngā Whare Taonga – Māori Treasures and European Museums’, Te Ara – the Encyclopedia of New Zealand. Retrieved 8 November 2021 from http://www.TeAra.govt.nz/en/interactive/44073/ rauru-wharenui-museum-fur-volkerkunde-hamburg. Te Rito, Joseph Selwyn. 2007. ‘Whakapapa: A Framework for Understanding Identity’, MAI Review 1(3): 1–10.

10

Travelling Houses Translation, Change and Ambivalence ♦l♦

A.-Chr. Engels-Schwarzpaul

Travelling Houses: Introduction Berlin, 2005: On the resort website, the Samoan Fale had been pictured sitting on a platform in the midst of the Tropical Village – its looks signalling the eternal sun and balmy breezes of the ‘South Seas’, that stereotypical wish image, to its mainly German and Polish visitors. The house’s carving, weaving and lashing details suggested other ways of life, where time moves at a different pace and simplicity harbours happiness. Now, on site, the fale is dwarfed by the massive steel structure of the former aircraft hangar that houses the Tropical Islands resort. Its handcrafted Pandanus mats and carved posts oddly contrast with the dome’s techno-feel. The atmosphere is loud and busy, even though there are not many visitors on this wintery Thursday afternoon. Actually, the fale looks deserted. London, 2012: Hinemihi is open today, surrounded by people – her people and half of London, it seems. It is the Ngāti Rānana Kōhanga Reo’s hāngī,1 and people are milling across the lawn in front of Clandon House in the morning sun. As I contemplate the strange contrast between Hinemihi and the huge oak tree above and watch the preparations, a karanga sounds, and suddenly the lawn transforms into a marae 2 ātea. Local and visiting Māori begin to stand in formation as they would at home

160  A.-Chr. Engels-Schwarzpaul



in Aotearoa. People become part of the pōwhiri and watch the performances before shopping at the stalls and getting some hāngī. Hamburg, 2014: Turning a corner on the museum’s top floor, I suddenly stand in front of Rauru. He looks very different from what I remember from the photos: a carefully arranged specimen, like an exquisitely lit jewel on black velvet. I had read that the presentation of the room was changed after his restoration, but I am not quite prepared for this: the house, small and beautiful, is surrounded by a simulation of New Zealand bush painted in garish colours. Weird. Even stranger, though, to walk down twelve steps towards a wharenui, approaching it from above and across a tiny open space. None of that had been visible in the video I saw of the 100th anniversary of Rauru’s arrival. Lā‘ie, 2014: We are too late for the performances. The stage for the reputedly hilarious performances of Samoan culture and the visitors’ benches seating crowds during the day are empty as we walk from the Māori towards the Samoan village. A big stone, like a Māori mauri, marks what I think must be the boundary, and beyond the malae beckons a huge faleafolau, with a smaller one on either side – all in an immaculate state, like most buildings here at the Polynesian Cultural Center. These ones, we hear later, were extensively restored just a couple of years ago.

My first encounters with those Māori and Samoan houses in Europe and in the USA3 already suggested how important relationships are for their status and roles. The houses and their current cultural, historical and geographical contexts shape each other. The resulting relational webs include original contexts and source communities; the local people who engage with them in their current locations; and, importantly, a global network of researchers, art buyers, exhibition agents and managers or commissioners of museums, theme parks and resorts. A bird’s-eye perspective focusing on global contexts therefore risks overlooking important local details, while a focus on a specific local context may miss global communalities. In this chapter, I shift between both, but ultimately lean towards a broad view, looking for communalities and shared patterns. In 2014, preparing for a visit to the Museum für Völkerkunde in Hamburg (today, Museum am Rothenbaum), I was wary of superficial analogies and nervous about how to explain my simultaneous interest in Rauru and the museum’s collections of historical photographs from German Völkerschauen (ethnographic shows). I need not have worried: for over a year, the museum had been showing two special exhibitions from its Oceania collection concurrently: ‘Blick ins Paradies’ (A glimpse into paradise) and ‘Südsee. Māori. Geheimnisse entdecken’ (South Seas. Māori. Discover secrets).4 This expansive mindset started with the Hamburg collections’ beginnings in the colonial era: Georg Thilenius, the museum’s



Translation, Change and Ambivalence  161

first full-time director, initially saw Rauru in New Zealand in early 1900. In 1908, he initiated, with his own ship, a South Seas expedition headed by Augustin Krämer (Neich 2001: 209; Triesch 2012: 196; Zwernemann and Wilpert 1990: 61). The expedition extracted thousands of artefacts from their original environments to reassemble them, ‘on lonely display’ (Bennett 2007: 129), as scientific objects in the collections. Their web of associations was and remains today reduced to the collections’ logic, with later efforts to reconnect them to their aura, to those who were gathered in them, unable to ‘strengthen their claim to reality’ (Latour 2004: 237; emphasis in original). As specimens, they became a ‘tangible representation of something “Pacific”’ (Mallon 2010: 27): things that appear very different in the Pacific can seem similar in the collection. On the other hand, a greater contextual rendering of details highlights similarities and differences in the Pacific that tend to disappear behind shared dissimilarities from European characteristics in Hamburg – thus, a specimen’s location impacts on the appreciation of its characteristics, value, and similarities or differences with others. When, from the second half of the twentieth century, Māori whare (houses) and Pacific fale (houses) travelled to far-away destinations, they changed from sacred sites, shelters, ancestral taonga (prized possession) and living houses to – depending on occasion and context – commodities, museum specimens, garden follies, theme park exhibits, transcultural meeting spaces, and temporary homes for diasporic communities. Neither one-directional nor final, such changes depend on the relationships of which the houses partake at any time. As isolated specimens on display, they may have no apparent connection to their source communities. Alternatively, they can be catalysts in collaborations between original guardians and overseas hosts or keepers, which change the relationships between source communities, legal owners and the houses themselves. One indicator of change is that institutions outside of Aotearoa increasingly address Māori whare whakairo (carved houses) or whare tūpuna (ancestral houses) by their ancestral name and the personal pronouns reflecting the gender of the ancestor: Hinemihi is ‘she’ and Rauru ‘he’. Before discussing the transformations arising from the houses’ travels in relation to intercultural translation processes, I will now provide some details about the journeys of two Māori and two Samoan houses to European and US exhibitions. Of particular interest here are relationships involving neighbours, kin, travellers, hosts, guests and ambassadors.5 Subsequently, I will explore notions of re-translation, ambivalence and change as factors impacting the conditions of possibility under which houses take on different roles.

162  A.-Chr. Engels-Schwarzpaul



Transformations: Travelling Histories Hinemihi o te Ao Tawhito, from Te Wairoa in Aotearoa New Zealand, was the first Māori wharenui (large house or meeting house) to arrive in Europe. She was commissioned by Aporo Te Wharekaniwha and carved by Wero Taroi and Tene Waitere, for the dual purposes of a whare tūpuna for Ngāti Hinemihi, the local hapū,6 and as a venue for entertainment, for locals and tourists. After a volcanic eruption in 1886, Ngāti Hinemihi abandoned the area. In 1892, the returning Governor of New Zealand, the Earl of Onslow, purchased the whare for £50 to take home to England as a memento and garden folly for Clandon Park, Surrey (Sully 2010). There, Māori Battalion soldiers rediscovered her during the First World War. In the 1980s, almost a century after her arrival, a large group from Ngāti Hinemihi came to gift and bless some newly created carvings, initiating a period of re-appropriation by Māori communities including Ngāti Rānana. Hinemihi has since regained public and spiritual presence, increasingly serving as ‘a Maori Ambassador at the centre of a transcultural partnership between British people and New Zealanders, Maori and non Maori’ (‘About us: Te Maru O Hinemihi’ n.d.). As a ‘National Trust property’ with a Grade II listing, she is also registered in a collection of historic buildings. These different roles often conflicted in the past, but they have also galvanized ‘Hinemihi’s people’ (Sully et al. 2014). Thus, Te Maru o Hinemihi7 lobbied to turn Hinemihi into a semi-functioning marae (a Māori communal meeting place) by adding sleeping facilities – in the process giving rise to collaborations with anthropologists and architects. But then a fire almost destroyed the Clandon Park mansion in April 2015. Hinemihi survived the incident, but any extension and even restoration was put on hold. Her people in the UK want the project to go ahead, her people in Aotearoa would like her returned.8 Rauru, by comparison, had a rather isolated existence for most of his time at the Museum für Völkerkunde. His origin story is unusual. Many of the carved panels were originally produced in the mid-nineteenth century by Te Waru, a local leader. Much later, after a series of misfortunes in the wake of a violation of tapu (restriction) during their production, Te Waru sold the panels to George Nelson, Rotorua hotelier and art dealer (Bennett 2007). Nelson then commissioned Tene Waitere9 to use these carvings in the construction of a new house. In 1900, this house, Rauru, was consecrated twice, by two tohunga (priests) from different tribes. Hirini Moko Mead (2003: 79) considers this a ‘highly irregular’ occurrence, suggesting that, in this way, Nelson intended to ‘authenticate the house as a traditional work’, raising Rauru’s market value as a Māori artefact (79; see also Ellis 2012; Triesch 2012).10 And, indeed, Nelson soon started protracted



Translation, Change and Ambivalence  163

negotiations with potential buyers. In 1904, Rauru was sold to an overseas buyer. Six years later, Thilenius bought Rauru for the Hamburg Museum from international art dealer Umlauff. Since 1912, Rauru has stood ‘transfixed’ (Parahi 2012) in the same custom-built gallery, yet his people flew around the world to visit him: during successive visits, Māori groups and individuals introduced different meanings and values into local curatorial processes. As a result, Rauru is now seen increasingly as an ‘ambassador of the Te Arawa people, and of Māori culture in Europe overall’ (Te Tenehi Teira, in Köpke 2012: 18). However, his day-to-day role at the museum is still primarily that of a specimen on display, and the approach towards the house is eerily peculiar.11 Yet, Rauru ensures that New Zealanders have free access to the museum and, for Māori, he provides a tūrangawaewae (place where one has rights of residence) and the right to sleep overnight.12 The Moata Samoa at the Polynesian Culture Center (PCC) in Lā‘ie, Hawai‘i, was built between 1961 and 1963 by matai tufuga (chief builder) Uga Muasau Alo of Faleniu (American Samoa). The Center had been initiated by Matthew Cowley, Latter Day Saint missionary, who proposed that Polynesian cultures and traditions would ‘endure if they were shared with others’ (Polynesian Cultural Center n.d.). He anticipated the day when ‘my Māori people’, just like the ‘Samoans and all those islanders of the sea’, ‘will have a little village … at Lā‘ie with a beautiful carved house’ (Polynesian Cultural Center n.d.). Tufuga Falefitu Masoe contributed the lashing and finishing work on the fale (pers. comm. between Albert Refiti and Delsa Atoa Moe, 15 October 2015), and much of the roofing was done by volunteer labour missionaries from Sāmoa (pers. comm. with Seamus Fitzgerald, 23 October 2015). Opinions about the fale’s integrity vary, and both the outer appearance of the buildings and the performances themselves have been criticized. Some Samoans argue that the performances exploit and commercialize Samoan culture, while others find the shows entertaining or ‘hilarious’: Parts of it reminded me of growing up in Samoa. I loved the umu, the smell, and being in a fale again. It’s been forever since I’ve done that. Just talking with the Samoan boys – I loved it [my friend, however,] hated it. … I was surprised we had such different views of the same thing when we are from the same place. (HS, performer at PCC in an interview, Mānoa, 16 September 2014)

The village serves, at least to some extent, as a home away from home to Samoans in Hawai‘i, and the Polynesian Center as a whole brings together communities from across the Pacific in collaborations.13 To this day, however, the Center negotiates considerable tensions between a community/educational mission, on one hand, and the requirements of a large

164  A.-Chr. Engels-Schwarzpaul



commercial enterprise, on the other. Whatever the assessment of their integrity, the Samoan fale at PCC are clearly not forgotten; they are connected to local Samoan communities, and those working in the Samoan village themselves have become a ‘very tight community’ who not only work but also travel together (HS, interview, 16 September 2014). This is not the case for a nameless fale at Tropical Islands resort close to Berlin, Germany. It originated in a business scenario, in which Malaysian investor, Colin Au, wanted to acquire houses that would appeal to the desires of what he imagined to be ‘the German market’ (Engels-Schwarzpaul and Simati Kumar 2014). Au travelled to Sāmoa to commission the production of a fale according to a photograph he supplied. Under the management of the Samoan Tourism Authority, several tufuga-faufale (expert builders) constructed the fale, assisted by a carving class at the local high school and several village communities providing materials. In 2004, the parts of the completed fale were shipped to Germany, where several tufuga reassembled it on site at the resort, as one of several ‘exotic’, ‘authentic’ houses in the Tropical Village. A mere commodity up to that point, the fale was later opened and blessed in the presence of Samoan officials. For some months, until November 2005, a Samoan troupe of eighty dancers and musicians performed close-by and in front of special guests hosted in the fale. For some, the fale was a home away from home, and they would occasionally meet and hang out there. During their engagement, many, to their surprise, learnt more about some aspects of their culture than they would have in Sāmoa. However, many felt isolated, suffered from the weather, and received significantly less than the minimal German wage. Samoan negotiators had hoped to build a lasting Samoan presence at the German resort, but the subsequent management saw no value in their presence. From 2006 onwards, tropically dressed-up local performers replaced Indigenous troupes.14 Soon, the fale sank into forgetfulness. By 2008, it was called Kalmoa Lounge and used as a part-time bar and smoking lounge, littered with empty glasses and cigarette butts. Visitors I asked at that time in close vicinity were completely unaware of a Samoan fale at the resort. For all these houses, critical factors determining their status, mode of being and function are whether or not the connections with their source communities remained intact throughout their many transformations, and whether new ones were established with the communities at their current locations. Therefore, I now want to consider a process that relies crucially on displacement, relational equations and change – namely, translation. Translations and re-translations (that is, the transfer and negotiation of meaning, function and relationships between different locations or communities) can, I propose, charge the houses with energy and enable them, in turn, to energize the relationships that are gathered in or around them.



Translation, Change and Ambivalence  165

Translations: Transforming Relationships ‘To translate is to displace’ (Callon 2007: 75), and to create new alignments and configurations. Etymologically, translating as ‘carrying across, removal, transporting’ (Harper 2001–2017) preserves a sense of spatial distance to be overcome, as well as of extraction and re-embedding of meaning and value. Taking this spatial dimension seriously, we can see how Pacific houses change in the transfer from the Pacific to new host countries. The meaning of translate as ‘remove from one place to another’ is attested from the fourteenth century (Hoad 2003), well before European exploration, conquest and colonization. Increasingly, to translate was to conceptualize and ‘express in one’s own language what others say and want, why they act in the way they do and how they associate with each other’ (Callon 2007: 75). Translating, in this sense, became crucial as colonizing Europeans encountered the acts, thoughts and associations of people elsewhere. In the ‘turn from one language to another’ (Hoad 2003), translation then considers, from a particular point of departure, the practices and thoughts onto which a language opens like a window.15 My own first significant experiences of cross-cultural conceptual translations involved the displacement of notions from European languages, specifically German, into Māori contexts and vice versa in the 1980s. To relate to a new world, in the wake of my own displacement from Germany to Aotearoa New Zealand, I had to make transpositions that went far beyond approximations of terms between different vocabularies. They ‘dishevelled’ my understanding of concepts, in a lasting way. Carl Mika describes this process for whakapapa (genealogy): The ground that emerges from whakapapa, and is important to Māori generally, is more powerful than its association with its common translation of ‘genealogy’ … whakapapa deals with the world on its own terms and can reinterpret academic dogma … whakapapa dishevels it in some way, or clears it even temporarily. (Mika 2016a: 48)

Since entering the long process of translation over thirty years ago, words and concepts have been inherently provisional for me. Aware of other possible ways of looking, placing and layering, I now look for relationships between terms: coincidence or contradiction between, for example, ways in which things are seen as important and durable, how they are liked and cared for, and how they are able to gather different forces (Latour 2008: 47–49).16 Translation, always itself a relationship, is theoretically an impossible task. Practically everyone is involved in it. Here in this chapter, translation

166  A.-Chr. Engels-Schwarzpaul



includes the interpretation and explanation that produce the houses’ histories, including my own efforts at making sense of and explicating such histories.17 Antoine Berman considers reflection on translation ‘an internal necessity of translation itself ’ (1992: 1), which here involves particularly the houses’ relational webs connecting their original contexts (their source communities) and the contexts of their current locations (their new communities, in an increasingly shared global context). Translation amounts to nothing if it is not a ‘putting in touch with’ (1992: 4). Therefore, as a first step in building relationships, I bring concepts and terms from different traditions together here, to engage the plurality of forces acting on the travelling houses in translation. In Aotearoa, writers like Hirini Moko Mead (2003) emphasize the importance of utu (reciprocity) and whakapapa (genealogy, layer, worlding, see Mika 2016b), reciprocal and intergenerational relationships underpinning thinking and practice at every level of Māori society. Bob Jahnke (1999) highlights how relationships between insiders and outsiders are articulated across the pae, a ‘transitional zone … of demarcation and negotiation’ (193–94).18 Samoan Head of State, Tui Atua Tupua Tamasese Efi discusses tua‘oi (neighbour, boundary) and vā fealoaloa‘i (relations based on mutual respect), in which separation and connection between self and other are shaped so that hospitality becomes possible (Tui Atua 2007). In Germany, Walter Benjamin invokes a ‘secret agreement between past generations and the present one’ (1969: 254), which commits us to recognize and redeem the past in the present. He insists on the dialogical nature of translation and the kinship of all languages and includes, like Pacific thinkers, non-human ones. Paul Ricœur (2006) in France establishes links between translation and hospitality, and Mario Erdheim (1992) in Switzerland unpicks an ambivalence between that which is one’s own and that which is other (dem Eigenen und dem Fremden) as potentiality in cross-cultural relationships. In the Pacific, Epeli Hau‘ofa (1994) compellingly appeals to the long tradition of Pacific peoples, travelling and enlarging their world through relational networks. Finally, Sean Mallon and Roger Neich argue (against modernist theories of meaning and aesthetic criticism) that cultural products manifest and mediate social and cultural processes, and their uses often transcend their appearance (Mallon 2010: 24). For more than a century, and across half the globe, the four houses discussed in this chapter have provided sites of engagement where transformative intercultural spatial production can take place – within relational networks involving kin, neighbours, hosts, guests, travellers and ambassadors. The acts of translation and retranslation accompanying their travels, between different locations and contexts, can not only be construed as operations ‘across systemic borders’; to translate (from Latin, tra-ducere)



Translation, Change and Ambivalence  167

is also to ‘conduct through, pass beyond, to the other side of a division or difference’ (Bal 2003: 6), adding a sense of intimacy and immersion. As a conceptual device, translation helps explore not only the changes in the houses’ effectivity but also their affectivity. When Māori whare and Samoan fale were transferred from their communities of origin to Europe and the US, one of their typical features, their ‘ability to stand in both worlds’ of noa (unrestricted/common) and tapu (restricted/sacred) (Wineera 2000: 25) took on a new inflection. Even in their new environments, their roles and functions still change with context and events and, notably, depending on the participants’ relationships with the houses. In Sāmoa, the term fale stands for shelter – so much so that a faletele (meeting house) cannot be conceived without the relationships it shelters, structures, expresses and relies on (Refiti 2015: 76, 101, 159, 250). In the Samoan nu‘u (village), notions like tua‘oi and vā fealoaloa‘i articulate connections with one’s own and separation from the other. Vā fealoaloa‘i relates ‘specifically to the relational bond between different entities’: people, their gods/God, and ‘the seas, skies and stars’ (Tui Atua 2007).19 To move beyond or to challenge, shift and fix tua‘oi is ‘part and parcel of defining and living human life. Different tua‘oi or boundaries mark the moving tides of culture and politics’ (Tui Atua 2007). These tides show the mutability of borders between one’s own and the other: relationships keep changing. Since translation, too, is an always unfinished process, new relations can be opened up endlessly. A story can always be told differently.

Transmutations: From Shelter to Specimen and Back When Hinemihi and Rauru were transferred to Europe, however, their relational context of kin and neighbours made way for relationships with strangers unfamiliar with the local idiom in Aotearoa. Widely shared colonial presumptions, such as Europe’s fundamental superiority, prevented them from generatively translating the houses’ modes of signification and use into their own milieu. Given their habitual formal-aesthetic or anthropological judgements and their lack of connection with the original communities, they could not contribute new dimensions to the interplay between kinship and mutual foreignness. Thus the houses became isolated and immobilized specimens, which were now placed into relationships governed by instrumental imperatives with other cultural representations (Kirshenblatt-Gimblett 1998: 165). Thus, Rauru was immediately isolated in a custom-built gallery in an ethnological museum; Hinemihi was displayed as a curio in a landscape garden and used as a summer play house and storage facility (Sully and Gallop 2007); and, later, the fale at Tropical Islands was turned into a bar and smoking lounge. While, from

168  A.-Chr. Engels-Schwarzpaul



one perspective, the Moata Samoa is also a specimen amongst other Pacific houses at PCC, it still serves its original dual purpose as home for diasporic Samoans in Hawai‘i and exhibit for paying tourist visitors – and this is perhaps a hint at a crucial factor determining success or failure of translation. Originally, Hinemihi also had a dual purpose – and as ancestor and tūrangawaewae for whakapapa and kaupapa whānau,20 as well as ambassador and host for other communities, she returned to such a role following her re-appropriation by Māori communities. What she will be in the future depends, inter alia, on the commitment and creativity of her people. Benjamin’s ‘secret agreement’ with past generations is easier to imagine locally, but Māori traditions of manaakitanga (hospitality) offer cross-cultural points of connection.21 Visiting and hosting train and expand the imagination by providing a distance from the known and familiar and create moments during which it becomes possible to look ‘upon the same world from one another’s standpoint’, seeing ‘the same in very different and frequently opposing aspects’ (Arendt 1961: 51). Without becoming the same, the different conditions of hosts and guests standing next to each other produce a common world (Peng 2008: 74).22 Collective, imaginative (re)translations go beyond understanding how other people ‘construct the world or are constructed by the world’ (Smith 1999: 37). A ‘way of sharing the world’ arises (Smith 1999: 37), which is oriented by mutuality and extends in many directions – if the translators’ respective languages are properly affected by the foreign. In Hinemihi’s case, a first translation inspired by nostalgic exoticism led to her change from whare tūpuna to garden folly and storage shed. As a memento, she also mediated ethnographic knowledge the Earl of Onslow had acquired as Governor of New Zealand (Fabian 2001: 120). However, translations are never final, and a re-translation process (led by Hinemihi’s people, ultimately including also the UK National Trust) led to her recognition as an ancestor and, subsequently, as ambassador. Embracing the material and spiritual agency of the whare, as well as a new global context of intensifying displacement, these changes gradually led to wider changes in conservation policies. Some years later, perhaps not surprisingly, the Hamburg Museum für Völkerkunde also recognized Rauru as an ancestor and as a European ambassador of Te Arawa and, more generally, of Māori culture (Te Tenehi Teira, in Köpke 2012: 18). Whereas collectors’ engagements with the houses have historically oscillated between desire and fascination with the strange and, on the other hand, fear and rejection (see Engels-Schwarzpaul and Refiti 2012; Engels-Schwarzpaul and Wikitera 2009a), many contemporary curators show a greater interest in the histories of the objects in their collections



Translation, Change and Ambivalence  169

and are willing to host and translate the strange into the local fabric – one of the fundamental conditions for cultural innovation (Erdheim 2002: 29). While lasting commonalities between cultures are important for stable relationships, the ambivalence and antagonism that sometimes arise out of chaotic encounters with the Other create productive challenges (Erdheim 1992: 743).

Change and Ambivalence: Global Re-Translations While a specimen and an ambassador both represent the groups they belong to outside of their contexts, agency marks the difference between a specimen’s paralysis and ambassadorial efficacy. Even though both may seem to have reached an endpoint on a continuum from communal taonga to isolated existence, recent scholarship dishevels assumptions of a linear and unidirectional development between sacred and profaned, taonga and commodity (e.g. Henare, Holbraad and Wastell 2007; Tomlinson and Tengan 2016; Wikitera 2015). In a 1993 interview with Paul Tapsell, Tomairangi Kameta emphasized the unpredictable nature of taonga: ‘taonga are our ancestors, they are like people, you can belong to them, but you can never own them. They have a freedom of their own. Some taonga become travellers. The old people use to say that eventually all taonga return home. I am inclined to agree. But no one can ever own them’ (Tapsell 1997: 343). Before their eventual return home, taonga can shift registers, stretched as they are across a tense field of unsettling ambiguity and ambivalence. In many European languages, the prefix ambi- signals the possibility of more than one: ambivalence (‘the coexistence … of mixed feelings or contradictory ideas’, Deverson and Kennedy 2005) and ambiguity (‘an expression able to be interpreted in more than one way’, ibid.) can be openings for different ways of knowing in multilateral translations – expanding, in this case, both Pacific and European vocabularies.23 This appears to have happened in the ‘Rauru Project’ leading up to the hundredth anniversary of his acquisition by the museum, in which staff at the Museum für Völkerkunde collaborated with Jim and Catherine Schuster, other Te Arawa and, more generally, Māori participants. The project necessitated a new vocabulary to support the parties in their collaborations. Director Wulf Köpke and Oceanic department curator Jeannette Kokott were surprised about the prevailing openness, as they and several other staff members formed previously unthinkable relationships with the Te Arawa community. The resulting whānau (extended family-like) relationships are expected to continue beyond the completion of the project: the museum is now part of Te Arawa history. The ‘ancestors’

170  A.-Chr. Engels-Schwarzpaul



in Hamburg are now a joint responsibility between the museum and Te Arawa: as caretaker, Köpke comments, the museum ‘must make sure that, for the next one hundred years, the meeting house continues to be well looked after and does not feel lonely again, and that the contact with the Te Arawa people will not be broken’ (Köpke 2012: 17). Te Tenehi Teira, National Māori Heritage Manager, emphatically advises that Rauru should be treated ‘like a member of the diplomatic corps’, so he can ‘create a bridge between the Māori culture and the German and European cultures and thus also secure the future of the Māori culture in New Zealand’ (Köpke 2012: 18). In my 2014 meetings and interviews with Kokott and her intern Marisol Fuchs, it seemed clear that both had taken on board, and were testing in their speaking and thinking, new ontological and epistemological possibilities.24 It was remarkable how Kokott and Fuchs held on to those different conceptions, back in wintery and rainy Hamburg where, historically and currently, houses are mostly artefacts and increasingly expensive real estate. These newly acquired understandings seemed to dishevel Albert Refiti’s proposition that one of the fundamental differences concerning relationships in Pacific and Western contexts is the prevailing type of exchange. The generative relationships Māori and Samoans looked for in the past, and still look for today, would lead to a rewriting of history (Benjamin 1969: 255) from a different perspective than that of the apparent or factual historical victors. At the Museum für Völkerkunde, such re-writing occurs with varying success since, intentionally or not, there is also ethnocentric resistance to translation, perhaps because of its potential to turn local culture around, as it were, and expose a previously enfolded side (Berman 1992: 7). In Western economies, objects are usually isolated from their contextual relationships to better circulate as commodities whereas, in the Pacific, the exchange of objects ideally strengthens the participants’ relationships. The disparity seems to be demonstrated in an interview with Köpke conducted for the Māori TV programme maraetv (Parahi 2012). Asked whether the museum would return Rauru if Te Arawa requested his repatriation, Köpke’s response is negative. On the other hand, he tries out Pacific ideas – at least rhetorically – when he accepts a caretaker role and insists that this question is ‘a spiritual matter’ and should not be reduced to a legal one. The different perspectives clash: ‘we are the caretakers of the house, … and so they would have to accuse us of neglecting that … the spirits of Rauru will make up their mind. If they want to stay here, they will stay, and if they decide that they won’t the house will go back’ (Köpke, in Parahi 2012: 4, 48–46, 21). Köpke uses here an English translation (caretaker) of the Māori term kaitiaki (or hunga tiaki), which implies a paradigm shift in the relationships



Translation, Change and Ambivalence  171

museums have with objects in their collections. In this shift, kaitiakitanga (guardianship) replaces ‘outright ownership’ and institutes mutual relationships and responsibilities (Tapsell 1997: 343, 362). When Köpke pronounces that ‘the spirits of Rauru will make up their mind’ about staying in Hamburg or going home, he seems to refer to taonga as travellers who ‘have a freedom of their own’ (Kameta, in Tapsell 1997: 343) and cannot be owned. However, ‘custodianship’, in Köpke’s dictionary (2012: 16), may well be synonymous with neoliberal notions of conservational ‘stewardship’ of disembedded artefacts. In ‘audit cultures’, with their ‘techniques of acquisition, research, and display’ and integral ‘desire to catalogue, conserve, and display objects’, such neoliberal notions generally do not replace, but rather justify legal ownership (Ouzman 2006: 269–71). Translation always involves interpretation – and when self-assurance and self-interest overshadow an interest in others’ thoughts, interpretations lean towards epistemic violence. Whenever museums’ desire to exercise control over the objects in their possession overshadows the interest in their history and source communities (see Hakiwai 2014: 11–20), misappropriation looms large. This problematic seems to reverberate in Köpke’s choice of words, ‘they would have to accuse us of neglecting [caretaker duties]’. It connotes adversarial legal battles rather than dialogical understanding or, for that matter, an appreciation that the relationship with taonga is never a proprietary one – rather, one belongs to them (McCarthy and Mané-Wheoki 2015: 314). When continued possession and control remains the primary concern, it is difficult to develop new forms of relationships. Assumptions about knowing the other in advance – in and on one’s own terms (Gadamer 1975: 360) – may then compound the objectifying effect of displays. Observers are then likely to ‘stand back from the artefact as object’ (Knappett 2014: 4703) and translate ‘the other into [their] own categories’ – capturing the others’ lexicon and grinding it up in their own conceptual machine – a certain menace to any dialogue (Caputo, in Kanaris and Doorley 2004: xi).

To the Other Side of Difference, Across Borders: Successful Translations By contrast, linguistic hospitality invites retranslations and implies endless engagement, balancing ‘the pleasure of dwelling in the other’s language … by the pleasure of receiving the foreign word at home, in one’s own welcoming house’ (Ricœur 2006: 10). Michael Goldsmith (2013) observes how the ‘dialogue between “local” and “foreign”’ mirrors the ‘cultural relationship between hosts and guests’ in Pacific contexts, demonstrating the contextual subtleties of relations between sacred Pacific buildings and their occupants.

172  A.-Chr. Engels-Schwarzpaul



These relational translations appear to hinge, for instance, on the sum total of the houses’ connections with their source communities. Hinemihi and Rauru were connected before their departure from Aotearoa and remain part of these networks to this day. By comparison, the Samoan houses were commissioned pieces with few local connections and instead were embedded in capitalist systems of exchange from the beginning. Another influential factor are Samoan and Māori worldviews more generally – particularly as they were impacted by colonial history. Whereas Western Samoans have considered themselves free and independent agents since 1962, Māori (like Hawaiian Kānaka Maoli) as First Nation or Fourth World people often experience their life worlds as overwhelmed by settler society. Consequently, the preservation, renewal and (re)definition of Māori culture are important aspects of historical and present resistance to assimilation pressures, and taonga are treasured, carefully guarded and protected. Travelling houses are known by name and route. Samoans, on the other hand, have developed a self-ironic commodification of cultural elements in response to global market demands (Balme 1998: 64), and Samoan travelling houses may cease to exist in local memory soon after their departure. As circulating objects cut loose from ‘their aura, their crown, their web of associations’, they could, in Latourian terms, be called failed gatherings (2004: 237, 246). When, however, houses are held in relational webs, they are ‘matters of concern’ (2004: 237), even as commodities, and can gather human and non-human participants in a relationally structured space. This is what the permanent exhibition, Tangata o le Moana: The Story of Pacific People in New Zealand (at the Museum of New Zealand Te Papa Tongarewa since 2007) wants to achieve by changing the interpretation and display of Pacific objects: the display is no longer about ‘examples of the exotic’ but about Pacific peoples as ‘neighbours, friends, spouses, extended family members, teammates and workmates’ (Mallon, in Bedford 2015: 215). Two phenomena have recently highlighted the global diversity of knowledges and interpretations: first, the exposure of an increasing number of people to diverse cultures in different parts of the globe; and secondly, the identity politics of Indigenous peoples. As a result, there is a growing realization amongst Western (and non-Western) scholars that all cultures, including their own, have blind spots; that each prohibits certain questions; and that all are unable to see their own blindness (Baecker 2012: 70, 109). This realization also affects how their keepers regard Hinemihi and Rauru – partly through a more intimate understanding of the ways in which colonization and globalization have directly and indirectly contributed to the changes that resulted for Hinemihi and Rauru from their very movements from place to place. Translation, in its sense of moving



Translation, Change and Ambivalence  173

from place to place, then appropriately mobilizes meaning and expands interpretive frameworks, facilitates multi-perspectival discussions, works around the blind spots in each culture and helps to uncover or articulate new affinities between Pacific and Western types of knowledge. In the past, museums exhibiting Pacific houses maintained their ownership partly by asserting their own rules over those of the houses’ source communities. Conflicts commonly arose from the tension between conservation requirements or curatorial strategies, on the one hand, and the associated communities’ spiritual, ceremonial or practical requirements, on the other. The New Museology movement has effected some change in orientation, giving people more ‘control over their cultural heritage and its preservation as part of how they maintain, reinforce, or construct their identity’ (Hakiwai 2012: 383). A source community’s ‘living culture and its continuing development’ (Christina Kreps, in Hakiwai 2012: 383) leads to a vision of shared ownership, for instance in recognition of the ‘real living relationships’ Māori have with their taonga (Hakiwai 2012: 393). As taonga, a whare whakairo can cement or establish relationships (in the past, its gifting marked important life events, brokered peace or called to war, see Ellis 2012: 426) which are then manifested in the current transactions between museums and source communities. When, however, the links with a source community are broken, as happened in many cases, restoring them is complex and relies on sometimes fraught processes of translation. A Māori colleague at AUT, for instance, who visited Hinemihi during her visit to London, experienced unexpected emotional turmoil (in the midst of otherwise overwhelmingly positive experiences) during a pōwhiri (ceremonial welcome). As a member of Hinemihi’s source community, she expected to participate on the side of the hosts, yet she was asked to join the party of guests to be welcomed – not necessarily an unusual thing to do, but a surprise in the situation, nevertheless. This experience caused her to explore and consider in greater depth the evolving cultural protocols in diasporic situations and how they relate to long-term, overarching Māori concepts (Wikitera 2015: 190, 217). Overwhelmingly, it seems, Māori accept such growing pains as part of a game in which much is to be gained. In line with international developments (Gurian 1999: 176, 181), an acceptance of kaitiakitanga and mutual responsibilities has thus replaced the outright assertion of legal ownership at the Museum für Völkerkunde in Hamburg (Köpke 2012: 17). The changing circumstances make it easier to think of a house’s change in status in more than one direction. While the idea of an irretrievable loss of authenticity is implicit in a sequence from sacred/ consecrated object, to specimen or artefact, to iconic object or souvenir, Deborah Root argues that the term ‘integrity’ indicates ‘a relatively unbroken connection between the image or object and the culture in which it is

174  A.-Chr. Engels-Schwarzpaul



made and used’ (Root 1996: 80). Such a connection does not rely on overt ownership and control to continue to shape the houses’ status and function. When the relationship changes, these changes are likely to change a house’s status as taonga, shelter, artefact, specimen or commodity – in the long or short term – in more than one direction. Mataatua Wharenui has clearly shown this. Built as a wedding gift by Ngāti Awa in Whakatane in 1874–1875, officially opened in 1875 and dedicated to Queen Victoria, Mataatua soon became a ‘focus for opposition to government land confiscation and purchase’ (Butts 2003: 98). Over the course of a century, the whare was then shipped off by the New Zealand government to exhibitions in Sydney and Melbourne, from there to the Victoria and Albert Museum in London, to be exhibited at the 1923 Wembley British Empire exhibition, then to be expedited to the South Seas Exhibition in Dunedin, after which the Otago Museum acquired it for its collection. There, Mataatua stood for decades in relative obscurity before coming to greater prominence during the 1986–1987 Te Māori exhibition after its return from the United States. [Mataatua] served as a venue for powhiri to welcome the huge crowds that attended the popular exhibition ... This ground-breaking exhibit ushered in sweeping changes in museum practice in terms of Maori involvement in the collecting and display of taonga Maori, leading ultimately to calls for consultation, co-management and even repatriation of several prominent houses, carvings and other ancestral treasures. (McCarthy and Mané-Wheoki 2015: 311)

In 1996, a settlement under the Treaty of Waitangi returned ‘the house that came home’ to Whakatane, where it was restored and reopened in 2011, and is now the core around which the activities of a cultural centre revolve. Another return seems to be imminent: in the aftermath of the fire, Hinemihi’s people negotiated a so far unprecedented agreement (see Hoete, Chapter 9 in this volume): the original carvings are to be returned to Aotearoa in exchange for new carvings (Boynton 2019). The exchange creates an opening for innovative collaboration, a process in which people living in both countries would kō-design,25 and co-produce a new whare in Clandon Park. Anthony Hoete, who has been involved with Te Maru o Hinemihi (a collective caring for Hinemihi; most members are Māori living in London but it also includes Jim and Cathy Schuster from Tuhourangi, Rotorua), imagines the new whare as Hinemihi o Te Ao Āpōpō (Hinemihi of the Future World). A 2017 (now somewhat dated) Te Maru proposal advocated Hinemihi’s future as a whare manaaki, a community house that would allow Ngāti Rānana to look after people overnight. The proposed



Translation, Change and Ambivalence  175

addition of cooking and eating areas, as well as a toilet, under the same roof might have been problematic in the original whare, given that tapu (restricted) and noa (common) spaces are normally separated on marae. However, the exchange project might now allow revisions of tikanga (protocol) in the new context. This, and several questions underpinning the design principles, remain to be negotiated, and the future whare at Clandon Park might be, just as Hinemihi o te Ao Tawhito was originally, a house that builds relationships in a changing world. A.-Chr. Engels-Schwarzpaul is Professor in Spatial Design and Postgraduate Studies at Auckland University of Technology – Te Wānanga Aronui o Tāmaki Makau Rau, in Auckland, Aotearoa/New Zealand. She was Executive Editor of  Interstices: Journal of Architecture and Related Arts  from 2005 to 2012  and edited approximatly eight special issues for  Interstices,  ACCESS and Knowledge Cultures. She also edited  Of Other Thoughts: Non-Traditional Ways to the Doctorate: A Guidebook for Candidates and Supervisors (2013, Sense Publishers, with M.A. Peters) and edited and translated  Atmospheric Architectures: The Aesthetics of Felt Spaces (2017, Bloomsbury, author Gernot Böhme).

Notes   1. Ngāti Rānana: the Māori expat community living in London; Kōhanga Reo: Language Nest (pre-school Māori language education); Hāngī: Māori method of cooking food using heated rocks buried in a pit oven. See Anthony Hoete’s chapter in this volume.   2. For a different account, see Bennett (2007: 135).   3. Other houses that travelled from the Pacific to Europe or America include: Mataatua, the first whare (house) to leave Aotearoa/New Zealand. Built in 1875 by Ngāti Awa in Whakatāne, the house was sent to an exhibition in Sydney in 1879, then purloined and expedited to London in 1881, sent back to Aotearoa only to be incorporated in the collections of the Otago Museum, and finally returned to Whakatāne in 1996. Hinemihi, Rauru und three other Māori houses are likely to stay in their current overseas locations. Ruatepupuke II from Tokomaru (opened 1881) was sold to Hindmarsh/Umlauff in the 1890s and then on to the current location at the Chicago Field Museum, arriving there in 1905. Te Wharepuni a Maui from Rotorua (opened prior to 1906) was hired out to the 1906–1907 New Zealand International Exhibition in Christchurch and sold in 1911 to Germany, where it is now held in the Linden Museum, Stuttgart. Te Aroha o Te Iwi Māori was commissioned by and built for the Polynesian Culture Center in 1963 (for a discussion of several of these whare, see Ellis 2012). The first known Samoan fale (house)

176  A.-Chr. Engels-Schwarzpaul

  4.

  5.

 6.   7.   8.

  9. 10.



to have travelled (allegedly previously owned by King Mata‘afa) was destined for the 1893 World’s Columbian Exhibition in Chicago. A fale from Mulinu‘u, commissioned by the NZ Government, travelled to the 1924 British Empire Exhibition at Wembley; and another fale was commissioned for the 1925 NZ and South Seas International Exhibition in Christchurch, New Zealand. In the 1970s, a cluster of Samoan fale were included in the Little World Museum of Man in Aichi, Japan, and, in 2004, a fale arrived at the Tropical Islands Resort in Brand, Germany. The collection covers a wide field: Sāmoa, Aotearoa and several other Pacific Islands (‘South Seas Treasures and Maori Art’), along with Australia, Micronesia, Melanesia, Malaysia, Indonesia and the Philippines. For more detailed accounts, see Bennett (2007), Ellis (2012), Engels-Schwarzpaul (2007a, 2007b, 2017), Engels-Schwarzpaul and Simati Kumar (2014), Engels-Schwarzpaul and Wikitera (2009b), Kokott (2012), Skinner (2008), Sully, Raymond and Hoete (2014), Te Maru o Hinemihi (http://www.hinemihi.co.uk), Wikitera (2015) and Wineera (2000). A hapū is a ‘large kinship group and [the] primary political unit in traditional Māori society’ (Moorfield 2017). Te Maru o Hinemihi is one of four project partners advancing future developments of Hinemihi o te Ao Tawhito and a new marae at Clandon Park. Ngāti Rānana hope that a growing partnership between Hinemihi and her source community, Ngāti Rānana and the UK National Trust will see Hinemihi ‘transformed from a vulnerable historic building into an active marae (Maori ceremonial space) and a cultural centre for Maori activities and learning in Britain’. See http://www.hinemihi.co.uk/page.php?id=24 (accessed 19 May 2022). They consider this ‘a fantastic way of creating a Māori space in Europe that could operate in a similar way to the cultural centre … in Hawaii’. On the other hand, her hapū back in Aotearoa would prefer her to be returned, so she can ‘be cared for by the people where she came from’ http://www. radionz.co.nz/news/te-manu-korihi/272802/new-opportunities-for-hinemihi (accessed 19 May 2022). For more details about Hinemihi’s history and current involvements, see (Engels-Schwarzpaul and Wikitera 2009b). Waitere was amongst the survivors sheltering in Hinemihi during the eruption. Nelson also instructed the carvers to remove all contemporary imagery from the panels purchased from Te Waru to make them look traditional (Neich 2001: 204). There seems to be a curious difference between whare whakairo and faletele: while six named whare whakairo are or were included in museums or museum-like collections overseas, I am not aware of any faletele in a permanent museum collection. Might this be related to a particular aspect of authenticity, namely consecration, which is a desirable attribute for collectors (Root 1996: 80–81)? As a performance of connection and ownership, consecration bestows authenticity, and serious collectors like museums



11.

12. 13. 14. 15.

16.

17.

18.

19.

20. 21.

Translation, Change and Ambivalence  177

would have taken consecration as an endorsement of an object’s genuineness. At the Tropical Islands resort, an echo of this established assessment might have been a blessing that was performed at its opening, see below. Adrian Bennett explains this with the lack of processional approach and with the violation of tapu implicit in an unannounced arrival and entry into a whare (2007: 130). In the 1980s, the museum was given the right to use the house for wedding ceremonies, provided the participants remove their shoes. For more information about the fale at PCC, see Aikau (2012), Wineera (2000), Stillman (2004) and Webb (1998). For more information about the fale at Tropical Islands, see Engels-Schwarzpaul and Simati Kumar (2014) and Engels-Schwarzpaul (2007a, 2007b). On the problematics of translation as part of the discipline of anthropology, see Salmond (2013). The window metaphor is perhaps too passive: languages may carve out ‘distinct worlds’ as opposed to ‘the same world with different labels attached’ (Sapir, in Salmond 2014: 170). Although the concept of translation certainly has linguistic origins in almost all versions of its use, I am here more interested in moments of transformation of thought and action. Translation is both a metaphor and more than a metaphor here: the concept of translation itself accounts for ‘the solid fact of the plurality and the diversity of languages’ (Ricœur 2006: 11). Jahnke also provides a salient caution regarding mis-translations of kinship terms like iwi, hapū and whānau, which severed connotations of regeneration and genealogy which amount to the imposition of Pākehā terms of reference (196). The boundaries between these terms are, themselves, conditional on relationships: depending on a speaker’s context, one employs vā fealoaloa‘i when speaking of one’s ‘own personal relationships or about those relationships close to them’; whereas, when speaking as a third party about the relationships of others, particularly of relationships the speaker is not familiar with, ‘the term tua‘oi is the more appropriate’ (Tui Atua 2007). Whakapapa whānau: group related by descent; Kaupapa whānau: group related by a common cause. These concepts – like many others – have no doubt changed their meaning since the nineteenth century, but important aspects appear to have remained constant. Manaakitanga is a highly developed practice centred on the marae, which functions on the basis of mana whenua, the mana of the local people. The marae ātea, the open place in front of the meeting house, stands in both worlds of tapu and noa – it is restricted when set aside for ceremonial purposes such as the welcome of visitors but is freely accessible at other times. John Terell, curator at the Field Museum in Chicago, which hosts another

178  A.-Chr. Engels-Schwarzpaul

22.

23.

24. 25.



Māori wharenui, Ruatepupuke II, considers the marae ‘New Zealand’s greatest gift to the world’ (Terell et al. 2007: 109). He believes that his own, American culture has a lot to learn from emulating marae orientations and protocols. Though missionary Cowley’s assertion, that Polynesian cultures had a better chance of surviving when they were shared with others, was paternalistic, Erdheim also observes, in a different context, that local cultures can be revitalized when they host strangers (Erdheim 1992: 735). Likewise, Sven Ouzman recounts Johannes Fabian’s notion of ‘Ecstatic’ Ethnography (as an experience of being outside oneself, transgressing one’s boundaries of self-containment and distance, 2001: 119), to discuss the fact that, sometimes, to see oneself clearly, ‘warts and all’ (2006: 271) requires a radical change of position and the adoption of another’s perspective. Interestingly, the term ambiguity appears to originate at the beginning of European colonization, more precisely in the 1520s (from Latin ambiguus, ‘having double meaning, shifting, changeable, doubtful’, derived from ambigere, ‘to dispute about’, but also ‘to wander’, from ambi-, ‘about’ + agere, ‘drive, lead, act’, Harper 2001–2017). The more contemporary term ambivalence, ‘simultaneous conflicting feelings’, was coined in 1910, at the height of European imperialism, by Swiss psychologist Eugen Bleuler (from Latin ambi-, ‘both’ + valentia, ‘strength’, Harper 2001–2017). The plurality and energy implicit in these terms, from the beginning, can shift perception from settled conventions to as yet unnoticed possibilities. There was no opportunity to talk to Köpke while he was director of the museum. See Anthony Hoete’s definition of kō-design in Chapter 9.

References ‘About us: Te Maru O Hinemihi’. n.d. Hinemihi. Retrieved 1 November 2021 from http://www.hinemihi.co.uk/page.php?id=24. Aikau, Hokulani. K. 2012. ‘More Than Preserving a Polynesian Paradise’, Arena Journal (37/38): 129. Arendt, Hannah. 1961. Between Past and Future: Six Exercises in Political Thought. New York: The Viking Press Inc. Baecker, Dirk. 2012. Wozu Kultur? Berlin: Kadmos. Bal, Mieke G. 2003. ‘Ecstatic Aesthetics: Metaphoring Bernini’, in Claire Farago and Robert Zwijnenberg (eds), Compelling Visuality: The Work of Art in and Out of History. Minneapolis: University of Minnesota Press, pp. 1–30. Balme, Christopher B. 1998. ‘Staging the Pacific: Framing Authenticity in Performances for Tourists at the Polynesian Cultural Center’, Theatre Journal 50(1 – Theatre, Diaspora, and the Politics of Home): 53–70.



Translation, Change and Ambivalence  179

Bedford, Richard. 2015. ‘Sean Mallon, Kolokesa Mahina-Tuai and Damon Salesa (Eds.) (2012) Tangata o le Moana. New Zealand and the People of the Pacific. Wellington: Te Papa Press [review]’, New Zealand Sociology 30(1): 215–31. Benjamin, Walter. 1969. ‘Theses on the Philosophy of History’, in H. Arendt (ed.), Illuminations. New York: Schocken Books, pp. 253–64. Bennett, Adrian John Te Piki Kotuku. 2007. ‘Marae : A Whakapapa of the Maori Marae’, PhD dissertation. Christchurch: University of Canterbury. Berman, Antoine. 1992. The Experience of the Foreign: Culture and Translation in Romantic Germany (trans. S. Heyvaert). Albany: State University of New York Press. Boynton, John. 2019. ‘Te Wharekaniwha Descendents “Kept in Dark” Over Return of Carvings’, Te Manu Korihi. Retrieved 2 November 2021 from https://www.rnz.co.nz/news/te-manu-korihi/404977/te-wharekaniwhadescendents-kept-in-dark-over-return-of-carvings. Butts, David. James. 2003. ‘Māori and Museums: The Politics of Indigenous Recognition’, PhD dissertation. Palmerston North, New Zealand: Massey University. Callon, Michel. 2007. ‘Some Elements of a Sociology of Translation: Domestication of the Scallops and the Fishermen of St. Brieuc Bay’, in Kristin Asdal, Brita Brenna and Ingunn Moser (eds), Technoscience: The Politics of Interventions. Oslo: Oslo Academic Press, Unipub, pp. 57–78. Deverson, Tony and Graeme Kennedy (eds). 2005. The New Zealand Oxford Dictionary. Oxford: Oxford University Press. Ellis, Ngarino. 2012. ‘“No Hea Koe? – Where are you From?” Māori Meeting Houses Overseas’, in B. Schmelz and W. Köpke (eds), The House Rauru – Masterpiece of the Māori. Mitteilungsband 44. Hamburg: Museum für Völkerkunde, pp. 419–35. Engels-Schwarzpaul, A.-Chr. 2007a. ‘Travel in Tropical Islands – Enemies Coexisting in Peace’, Interstices: Journal of Architecture and Related Arts 8: 21–30. . 2007b. ‘“A Warm Grey Fabric Lined on the Inside with the Most Lustrous and Colourful of Silks”: Dreams of Airships and Tropical Islands’, The Journal of Architecture 12(5): 525–42. . 2017. ‘Performing Diasporic Relationships: Pacific Houses in Europe’, The Contemporary Pacific 29(1): 38–64. Engels-Schwarzpaul, A.-Chr. and Albert L. Refiti. 2012. ‘Fale Samoa and Europe’s Extended Boundaries: Performing Place and Identity’, in Hilde Heynen and Janina Gosseye (eds), EAHN – Proceedings of the 2nd International Conference of the European Architectural History Network. Brussels: EAHN European Architectural History Network, pp. 337–44. Engels-Schwarzpaul, A.-Chr. and Benita Simati Kumar. 2014. ‘A Fale Samoa at Tropical Islands Resort, Germany: Performing Samoa to the World’, Tracing

180  A.-Chr. Engels-Schwarzpaul



Footprints of Tomorrow: Past Lessons, Present Stories, Future Lives. Proceedings of Samoa Conference II, 4–8 July 2011. Apia: NUS. Engels-Schwarzpaul, A.-Chr. and Keri-Anne Wikitera. 2009a. ‘Counter Currents: Whare Nui and Fale Abroad’, The Pacific Connection – Trade, Travel & Technology Transfer, 38–46. Retrieved 21 April 2022 from https://openrepository.aut.ac.nz/handle/10292/3677. Engels-Schwarzpaul, A.-C. and Keri-Anne Wikitera. 2009b. ‘Take Me Away … In Search of Original Dwelling’, Interstices: Journal of Architecture and Related Arts 10 Adam’s House in Paradise 42–54. Erdheim, Mario. 1992. ‘Das Eigene und das Fremde: Über ethnische Identität (Native and Alien: On Ethnic Identity)’, Psyche: Zeitschrift für Psychoanalyse und ihre Anwendungen 46(8): 730–44. . 2002. ‘Verzerrungen des Fremden in der Psychoanalytischen Perspektive’, in Ortrud Gutjahr (ed.), Fremde: Freiburger literaturpsychologische Gespräche. Jahrbuch für Literatur und Psychoanalyse Bd (Vol. 21). Würzburg: Königshausen & Neumann, pp. 21–45. Fabian, Johannes. 2001. Anthropology with an Attitude: Critical Essays. Stanford, CA: Stanford University Press. Gadamer, Hans-Georg. 1975. Truth and Method. New York: The Continuum Publishing Corporation. Goldsmith, Michael. 2013. ‘Dialogical Relations in and Between Sacred Pacific Buildings and Their Occupants’, in Albert L. Refiti (ed.), Pacific Spaces and Sacred Houses Fono #1, Auckland: Interstices Under Construction Symposium, pp. 4. Gurian, Elaine Heumann. 1999. ‘What is the Object of this Exercise? A Meandering Exploration of the Many Meanings of Objects in Museums’, Daedalus 128(3): 163–83. Hakiwai, Arapata. 2012. ‘“Honour the Future by Caring for the Past”: The Bicultural Mission of New Zealand’s National Museum Te Papa Tongarewa’, in Bernd Schmelz and Wulf Köpke (eds), The House Rauru – Masterpiece of the Māori. Mitteilungsband 44. Hamburg: Museum für Völkerkunde, pp. 379–97. . 2014. ‘He Mana Taonga, He Mana Tangata: Māori Taonga and the Politics of Māori Tribal Identity and Development’, PhD dissertation. Wellington: Victoria University. Harper, Douglas. 2001–2017. Online Etymology Dictionary. Retrieved 21 April 2022 from http://etymonline.com/ Hau’ofa, Epeli. 1994. ‘Our Sea of Islands’, Contemporary Pacific 6(1): 147–61. Henare, Amiria J.M., Martin Holbraad and Sari Wastell. 2007. Thinking Through Things: Theorising Artefacts in Ethnographic Perspective. Milton Park; New York: Routledge.



Translation, Change and Ambivalence  181

Hoad, T.F. (ed.). 2003. The Concise Oxford Dictionary of English Etymology. Retrieved 2 November 2021 from http://www.oxfordreference.com/ view/10.1093/acref/9780192830982.001.0001/acref-9780192830982. Jahnke, Robert. H.G. 1999. ‘Voices Beyond the Pae’, in Nicholas Thomas and Diane Losche (eds), Double Vision: Art Histories and Colonial Histories in the Pacific. Cambridge: Cambridge University, pp. 193–209. Kanaris, Jim and Mark J. Doorley. 2004. In Deference to the Other: Lonergan and Contemporary Continental Thought. Albany: State University of New York Press. Kirshenblatt-Gimblett, Barbara. 1998. Destination Culture: Tourism, Museums, and Heritage. Berkeley: University of California Press. Knappett, Carl. 2014. ‘Materiality in Archaeological Theory’, in Claire Smith (ed.), Encyclopedia of Global Archaeology (Vol. 7 (L-Mn)). New York: Springer Reference, pp. 4700–4708. Kokott, Jeanette. 2012. ‘One Hundred Years of Rauru in Hamburg: Building Bridges between Aotearoa/New Zealand and Germany’, in Bernd Schmelz and Wulf Köpke (eds), The House Rauru – Masterpiece of the Māori. Mitteilungsband 44. Hamburg: Museum für Völkerkunde, pp. 26–37. Köpke, Wulf. 2012. ‘More than just a Meeting House ...’, in Bernd Schmelz and Wulf Köpke (eds), The House Rauru – Masterpiece of the Māori. Mitteilungsband 44. Hamburg: Museum für Völkerkunde, pp. 13–25. Latour, Bruno. 2004. ‘Why Has Critique Run out of Steam? From Matters of Fact to Matters of Concern’, Critical Inquiry 30(2): 225–48. . 2008. What is the Style of Matters of Concern? Two Lectures in Empirical Philosophy. Assen: Koninklijke Van Gorcum. Mallon, Sean. 2010. ‘Beyond the Paperskin’, in Maud Page, Sean Mallon and Imelda Miller (eds), ‘Paperskin: The Art of Tapa Cloth’. Exhibition Catalogue (18 June 2010 – 26 September 2010). Wellington: Museum of New Zealand Te Papa Tongarewa, pp. 23–31. McCarthy, Conal and Jonathon Mané-Wheoki. 2015. ‘Afterword: The Wharenui Mataatua, and Some Thoughts About Things’, in Annabel Cooper, Lachy Paterson and Angela Wanhalla (eds), The Lives of Colonial Objects. Dunedin: Otago University Press, pp. 311–20. Mead, Hirini Moko. 2003. Tikanga Maori: Living by Maori Values. Wellington: Huia Publishers. Mika, Carl. T.H. 2016a. ‘“Papatūānuku/Papa:” Some Thoughts On The Oppositional Grounds Of The Doctoral Experience’, Knowledge Cultures 4(1): 43–55. . 2016b. ‘Worlded Object and its Presentation: A Maori Philosophy of Language’, AlterNative: An International Journal of Indigenous Peoples AlterNative: An International Journal of Indigenous Peoples 12(2): 165–76. Moorfield, John C. 2017. Te Aka Online Māori Dictionary. Retrieved 21 April 2022 from http://maoridictionary.co.nz/.

182  A.-Chr. Engels-Schwarzpaul



Neich, Roger. 2001. Carved Histories: Rotorua Ngāti Tarawhai Woodcarving. Auckland: Auckland University Press. Ouzman, Sven. 2006. ‘The Beauty of Letting Go: Fragmentary Museums and Archaeologies of Archive’, in Elizabeth Edwards, Chris Gosden and Ruth Phillips (eds), Sensible Objects Colonialism, Museums and Material Culture. Oxford: Berg, pp. 269–301. Parahi, Carmen. 2012. ‘Rāuru Wharenui Celebrates 100 years’, Maraetv. Retrieved 2 November 2021 from https://www.youtube.com/watch?v=dHG47c5ViHI. Peng, Yun. 2008. ‘A U-turn in the Desert: Figures and Motifs of the Chinese Nineteen Eighties’, PhD dissertation. Minneapolis, University of Minnesota. Polynesian Cultural Center. n.d. Early History | Polynesian Cultural Center. Retrieved 2 November 2021 from http://www.polynesia.com/early-history. html. Refiti, Albert. L. 2015. ‘Mavae and Tofiga: Spatial Exposition of the Samoan Cosmogony and Architecture’, PhD dissertation. Auckland, Auckland University of Technology. Ricœur, Paul. 2006. On Translation. Abingdon: Routledge. Root, Deborah. 1996. Cannibal Culture: Art, Appropriation, and the Commodification of Difference. Boulder, CO: Westview Press. Salmond, Amiria J. 2013. ‘Transforming Translations (part I) “The Owner of These Bones”’, HAU: Journal of Ethnographic Theory 3(3): 1–32. . 2014. ‘Transforming Translations (part 2) Addressing Ontological Alterity’, HAU: Journal of Ethnographic Theory 4(1): 155–87. Skinner, Damian. 2008. The Carver and the Artist: Maori Art in the Twentieth Century. Auckland: Auckland University Press. Smith, Linda Tuhiwai. 1999. Decolonizing Methodologies: Research and Indigenous Peoples. Dunedin: University of Otago Press. Stillman, Amy Ku‘leialoha. 2004. ‘Pacific-ing Asian Pacific American History’, Journal of Asian American Studies 7(3): 241–70. Sully, Dean. 2010. ‘Hinemihi, Clandon Park, Surrey UK’, A Brief History in Pictures. Retrieved 2 November 2021 from http://www.hinemihi.co.uk/media/121210-1209-history%20in%20pictures.pdf. Sully, Dean and Alan Gallop. 2007. ‘Introducing Hinemihi’, in Dean Sully (ed.), Decolonising Conservation: Caring for Maori Meeting Houses Outside New Zealand. Walnut Creek, CA: Left Coast Press, pp. 127–48. Sully, Dean, Rosanna Raymond and Anthony Hoete. 2014. ‘Locating Hinemihi’s People’, Journal of Material Culture 19(2): 209–29. Tapsell, Paul. 1997. ‘The Flight of Pareraututu: An Investigation of Taonga from a Tribal Perspective’, The Journal of the Polynesian Society 106(4): 323–74. Terrell, John Edward, Desiree C.J. Wisse and Christopher J. Philipp. 2007. ‘Ruatepupuke II, The Field Museum, Chicago: The Past and Possible Futures’,



Translation, Change and Ambivalence  183

in Dean Sully (ed.), Decolonising Conservation: Caring for Maori Meeting Houses Outside New Zealand. Walnut Creek, CA: Left Coast Press, pp. 89–109. Tomlinson, Matt and Ty Kawika P. Tengan. 2016. New Mana: Transformations of a Classic Concept in Pacific Languages and Cultures. Acton, ACT: ANU Press. Triesch, Carl. 2012. ‘From Rotorua to Hamburg: Rauru Finds a New Home’, in Bernd Schmelz and Wulf Köpke (eds), The House Rauru – Masterpiece of the Māori. Mitteilungsband 44. Hamburg: Museum für Völkerkunde, pp. 191–223. Tui Atua, T.T.T.E. 2007. ‘Samoan Jurisprudence and the Samoan Lands and Titles Court: The Perspective of a Litigant’, Public Lecture, 29 October 2007. Honolulu: University of Hawaii, Mānoa. Retrieved 21 April 2022 from http:// www.head-of-state-samoa.ws/pages/speech_jurisprudence.html. Webb, Terry D. 1998. ‘A New Kind of Plantation: The Polynesian Cultural Center in Lā`ie, Hawai‘i’, CRM (8): 33–36. Wikitera, Keri-Anne. 2015. ‘Māori Spaces in Foreign Places: Hinemihi o Te Ao Tawhito’, PhD dissertation. Auckland: Auckland University of Technology. Wineera, Vernice. 2000. ‘Selves and Others: A Study of Reflexivity and the Representation of Culture in Touristic Display at the Polynesian Cultural Center, Laie, Hawaii’, PhD dissertation. Honolulu: University of Hawai‘i, Honolulu. Zwernemann, Jurgen and Clara B. Wilpert. 1990. ‘The Hamburgisches Museum für Völkerkunde and Its Pacific Department: A Short History’, Pacific Arts: 60–62.

Conclusion

Vā What Is In-Between Architecture and Anthropology? ♦l♦

Albert L. Refiti and A.-Chr. Engels-Schwarzpaul

T

he questions raised by the exchanges between architecture and anthropology that inspired the publication of this edited collection went beyond our initial interest in the role of Indigenous Pacific concepts in the creation and maintenance of space. As architects, anthropologists, historians and theorists, contributors here address ecological aspects of Pacific spaces; relationships between tradition and creativity; personal exploration and wayfinding; cross-cultural global encounters or the dynamics of in- and exclusion (manifest particularly in threshold situations, in relative openness or closure); as well as fluidity and control. Their chapters provide external perspectives on the contents of others and, at the same time, show coherent transdisciplinary concerns – particularly at the intersection between anthropology and architecture. While these disciplines share common sources, for instance through the regular appropriation of philosophical concepts, architects and anthropologists often deploy ideas and terms in different ways. Such divergent uses of terminologies do not exactly act as metaphors for each other – yet, a disorientation caused by their juxtaposition, and thereby the deviation from a singular base-line (Cheng 2016), provides a space of ambiguity where new shades of meanings arise along with new questions. Looking at the same thing from different perspectives allows a worldly reality to arise that can unfold in different



Conclusion  185

ways (Arendt 1958/1998: 57). When conventional meaning is unsettled or ruptured, symbolic systems of space and relationality can ‘“make” and “remake” reality’ (Ricoeur 1978: 152) and multiply our views and descriptions of likeness and difference. Anthropologist Eduardo Viveiros de Castro coined the term ‘perspectivism’ to explain these different ways of meeting and seeing, suggesting that Western thought must go through an ‘exchange of perspectives’ to encounter at ‘right angles’ the thoughts of others (Viveiros de Castro 2012: 45, 77). At the outset of this project, we wanted to take up the threads of a discussion connecting architecture and anthropology that had started in the 1960s (Rapoport 1991; Rudofsky 1987; Rykwert 1988). A decade or two later, it more or less disappeared as the concern with embodied human experience was dissipated in the textual play of deconstruction, if not discredited by anti-phenomenological and anti-anthropocentric attitudes (Eisenman 1984). This discussion was rekindled by Tim Ingold in Making (2013) and, around the same time, Victor Buchli (2013; see also Yaneva 2009). However, its threads were still waiting to be taken up in specific local contexts. The collection presented here does this in and across different locations around the Pacific,1 closely connected to these locations and with Indigenous Pacific researchers in the forefront. A project that went through several stages since 2013, in order to find its current focus on architecture and spatial practices in the Pacific, is in danger of covering too wide a scope to fit into a single edited collection. And so it was: several lines of enquiry went off in different directions. The texts collected here have, through the specific connections they forge between anthropology and architecture, allowed us to think again about how social systems play out in architecture, and how important it is to reorient architectural research in this direction. Some contextual conditions left their traces, too. For instance, the research in this book occurred alongside work in the Vā Moana - Pacific Spaces research cluster, in which we engage both Pacific and Western thought to explore and understand contemporary and customary Indigenous ways of knowing and making the world – to produce neighbourhoods, spaces, objects, rituals and performances (Engels-Schwarzpaul and Mika 2019; Lopesi 2020; Lythberg 2020; Mahina 2019; Refiti et al. 2021; Refiti, Hoskins and Engels-Schwarzpaul 2021). This work has impacted on editorial strategies and the selection of chapters included here. The texts pivot out from Tāmaki Makaurau (Auckland) in New Zealand, into the wider Moana, and refract Western knowledges. We wanted first to make it relevant in our local situations and the relationships that matter here, to then identify elements and aspects that might be significant in global contexts. For what seemed at first as a specifically Indigenous concern has

186  Albert L. Refiti and A.-Chr. Engels-Schwarzpaul



more recently moved into general focus (at least in a formal sense, and in certain places). We registered a heightened interest in the inner workings of buildings regarding rituals, processes and connections with local communities. In modernist, post- and neo-modernist projects, the deployment of Indigenous (here in Auckland, Māori) characteristics was reduced to language or formal aesthetics, like surface add-ons. Current practice here (at least in the public sector) has moved to spatial analysis and planning models guided by Māori and Pacific principles. This is the case for the renovation of Tāmaki Paenga Hira Auckland War Memorial Museum,2 as well as in museums and cultural centres throughout the Pacific. Further examples include Renzo Piano’s monumental Jean-Marie Tjibaou Cultural Centre in New Caledonia (Brown 2002; Losche 2003; Sarvimäki 2019), the National Museum of Australia in Canberra, Te Papa Tongarewa Museum of New Zealand (both discussed in Walker 2018), and the much more modest buildings at the Hihiaua Cultural Centre in Whangarei, Aotearoa New Zealand.3 Since the 1990s, several wharenui and Pacific fale have been built at public institutions in Aotearoa (see Refiti, Hoskins and EngelsSchwarzpaul 2021) or in Hawai‘i, for instance the Hawai‘inuiākea School of Hawaiian Knowledge at University of Hawai‘i, Mānoa. Involving Indigenous experts or commissioning Indigenous artists to produce site specific work in such projects entails the establishment of different ways of working together, which, in turn, build relationships in the process. Often in such collaborations, naming is used to complement and extend architectural strategies into realms which architecture alone is unable to reach. These names, then, can become organizing concepts in the design process, as they did at Tāmaki Paenga Hira, where distinctive entry experiences acknowledge, respectively, the museum’s history and global heritage and twenty-first century culture(s) in Tāmaki Makaurau and Aotearoa. Ancestral relationships and the role of humans in birthing the world link both parts of the museum. Sites of cultural encounters are planned and implemented with varying degrees of involvement, recognition and monetary appreciation of the Indigenous communities they reference. As the model of cultural encounter appears to replace that of cultural difference (Christiansen, Galal and Hvenegård-Lassen 2017), forms of exchange become ever more varied, and the assessment of a project’s potential and subsequent success (or otherwise) more complex. The integration of an anthropological lens in such analyses, particularly in the hands of Indigenous experts, has much to offer here.4 Accordingly, the recognition of fruitful cross-overs between creative and anthropological endeavours manifests itself increasingly in collaborations between designers and anthropologists, for instance in the EcoDesignCollective’s initiatives.5 What once appeared to be the specific



Conclusion  187

concerns of local Indigenous movements and disciplinary experts is now a much more global and widespread engagement. One indication of this development is the increasing interest of Pacific artists, architects, educationalists, health experts and policy makers (to name but a few) in the Samoan and Tongan concept of vā. Commonly glossed as relational space, it refers to aspects of living together in relationships that include, but far exceed, architectural concerns. Vā plays an important part in understanding socio-spatial cultural systems in the Pacific; how these, in turn, structure social spaces; and how the latter are concretized with architectonics. To begin a dialogue between architecture and anthropology in the Pacific and beyond, then, means also looking afresh at technology and tectonics. Apart from Athol Greentree’s chapter, these aspects are beyond the scope of this volume, but we do want to sketch out a brief account and interpretation of Pacific notions about space. In Tonga and Sāmoa (though in slightly different inflections), vā refers to an interval, the relation between entities. For humans, this means a kinship system. In terms of English translation, the Samoan and Tongan vā is mostly rendered in spatial terms whereas wā, Māori, is rendered predominantly as temporal. However, these differences might partially be caused by the translation itself (somewhat paradoxically, since the division of time and space in English or German is anything but complete). Poet and essayist Jennifer Cheng speculates that translation, as an inherent part of language, is like refraction: ‘never a perfect transference of concepts but instead a mediated articulation of the world’ (2016). Linda Tuhiwai Smith maintains that ‘the Māori word for time or space is the same. Other Indigenous languages have no related word for either space or time, having instead a series of very precise terms for parts of these ideas, or for relationships between the idea and something else in the environment’ (Smith 2012: 52).6 This has implications for thinking about the concepts underpinning an architectural project, and how it addresses relationality – both in terms of human relationships in time and space and with regard to the objects that architecture creates in time and space. The seminal definition of vā was provided by Albert Wendt, who defined it, almost in passing, as ‘the space between, the betweenness, not empty space, not space that separates but space that relates, that holds separate entities and things together in the Unity-that-is-All, the space that is context, giving meaning to things’ (1996). He underscored the importance of vā for communal cultures, where a ‘person/creature/thing’ is perceived in relation to the group. Micah van der Ryn (2012) discusses relationships with Samoan spatial and architectural order in terms of point-field relationships. He uses Lehman and Herdrich’s definition of a point-field space as:

188  Albert L. Refiti and A.-Chr. Engels-Schwarzpaul



the topological neighbourhood of a given point, … Thus, any such point field is infinite, save as it ‘comes up against’ the field of a competing initial point, while all other points are understood (‘located’) as in one or other (or indeed both) fields, though each, in its respective if subordinate way, itself establishes a field, and so on recursively. (2002: 181; emphasis in original)

If represented as a series of vectors emanating from a point outward, this field extends indefinitely, and its boundaries are really relationships between points that are always contested and the focus of attention (2002: 184). The academic literature about vā/wā as a social, relational space has proliferated in the last decade or so. Most publications arise from Pacific groups’ experiences of diasporic constellations in Aotearoa New Zealand, Australia, Fiji and the USA and are motivated by a desire to build communities. In urban situations, where Pacific immigrants and Māori internal migrants often form alliances to gain leverage vis-à-vis the dominant political forces, the deployment of terms such as vā and wā across different Māori and Pacific cultures can come at the price of conflating distinct aspects into a generalized ‘Māori and Pasifika’ approach. The notable cross-referencing of sources between Māori and Pacific authors striving to invoke a vā in diaspora, in order to inspire and shore up Pacific futures, is therefore not without its own problems. Yet, it simultaneously indicates the usefulness of the concept. Vā emphasizes the importance of socio-spatial relationships when these, and their underpinning principles, can no longer be taken for granted: Pacific people in Aotearoa have no local ancestral lands to refer to as a matter of course. In cities like Auckland or Wellington, urban marae (Māori community spaces) have been developed by and for the majority of Māori whose tūranga waewae or ancestral marae is further away. They articulate new-old kinds of connections between land and people (Tapsell 2002: 142–43) and generate constellations in time and space in which things, people and forces can be properly aligned. From the 1970s, Pacific immigrants and their descendants in Aotearoa initially adapted existing community spaces and churches to establish new-old relationships in the cities. In a more distributed manner, Samoans in South Auckland also turned garages and backyards into spaces serving temporarily various functions that could not be accommodated elsewhere: the reception of visitors, funerals and ‘ava (kava) ceremonies, ceremonial exchanges and other formal activities, particularly fono (meetings; MacPherson 1997: 163). Reinventing ancestral cultures in new locations (Bedford 2001: 2), they adapted the layout and atmospheres of vernacular spaces to the occasions (Engels-Schwarzpaul 2019). Writing in Kona, Hawai‘i, Epeli Hau‘ofa (who was born in Papua New Guinea to Tongan parents and spent most of his academic life in Suva, Fiji) emphasized already in the early 1990s the



Conclusion  189

importance of ‘establishing new resource bases and expanded networks for circulation’ in a process of ‘world enlargement’ (Hau‘ofa 1994: 156, 151), but also of forging alliances with tangata whenua (people of the land) of Aotearoa and kānaka maoli (Indigenous people) in Hawai‘i (156). For in the often-precarious conditions of diaspora, alliances and well-functioning networks of relationships are vital. In their homelands, Samoan and Tongan interpellations like ‘ia teu le vā’ or ‘tauhi vā’ imply discord or danger – disturbed or disrupted relations that need to be healed or tended to, or even a state that is inherently culturally dangerous. But how does one tauhi or teu le vā in radically different environments, in which not even the tectonics of buildings materialize vā relationships in any permanent or dignified sense? Vā relationships are difficult to enact in European building types, which are organized according to principles that are very different from point-field spatiality. The latter is materialized in open Samoan fale where, for instance, the posts not only ‘hold the structural significance of supporting the large dome shaped roof ’; rather, ‘that significance becomes social metaphor when there are formal meetings in the structure and people sit according to their status and rank at different posts’ (Van der Ryn 2012: 139). This is one of the reasons why buildings like the Fale Pasifika at the University of Auckland play an important role in community development: forces operating in vā relationships find material expression in its design and construction. However, some other important aspects of fale building are absent there,7 most importantly the contributions of the main knowledge holders in Pacific Indigenous architecture, the tufuga/tufunga/ tohunga (master builders). They were replaced by engineers, architects, an advisory board and a Samoan architectural consultant. Without these Indigenous experts, though, cultural, ritual and technological knowledge that is integrated into traditional Pacific building projects could play no part in the Fale Pasifika’s construction. The building of Pacific fale in the diaspora has thus become the domain of builders and contractors – in contrast to another branch of traditional technological Pacific knowledge: waka (ship) building and sailing. In that area, Indigenous knowledge is again being taught in Aotearoa, most famously at Sir Hekenukumai Ngāiwi Puhipi (Hek) Busby’s Kupe Waka Centre at Aurere.8 The centre was designed with core principles such as tapu (restricted) and noa (unrestricted) as structuring devices around which architecture could wrap itself; conceptually, if not visually, this is akin to early Samoan communities’ repurposing of garages in South Auckland. Another reason why Māori and Pacific buildings, by bringing about relational connections through vā and generating wellbeing, have united diasporic communities in Aotearoa New Zealand since the 1970s is their

190  Albert L. Refiti and A.-Chr. Engels-Schwarzpaul



iconicity. Over 2000 years of Pacific migrations, climatic and geopolitical environments, together with cosmologies and ceremonies, have engendered ancestral house forms. Defying pressures to rationalize and adapt to new building forms, even largely modern buildings in Auckland’s suburbs, for instance, show off apse-like features of Tongan and Samoan buildings. These forms, once considered outdated or even obsolete, have not only survived but are now deployed by communities, businesses and institutions to grow identification and participation by Pacific people and tangata whenua. So, why do core concepts of built forms, originated during the migration into the Eastern Pacific, persist in urban diasporas? How do Pacific building traditions re-enact socio-spatial relationships and signify the values of kinship and rituals – in migrating, settling and enduring? Under which conditions do Pacific building forms create relational connections through vā (spatial relationships and practices) and become important sites of mana that shape and strengthen communities in the absence of ancestral lands? To answer these questions, and to understand both past and present processes involved, members of the Vā Moana - Pacific Spaces research cluster, under the leadership of Albert L. Refiti and Rau Hoskins, are presently embarking on a new project, ‘Artefacts of Relation: Building in the Pacific’. It starts from the premise that traditional Pacific buildings encode an understanding of belonging beyond the short-term, individual focus of mainstream Western planning. As we planned this new project, and while editing this collection, we collectively experienced the impact of Covid-19. Not only did the pandemic heighten our sense of the local and the global, but we also observed a shift in relational constellations. Different connections and disconnections across space and time made some collaborations closer and disrupted others – even though digital technologies (and the new vā relations produced by them) helped to maintain many relationships and functions. Finding ourselves in a new kind of shared space, without the solidity of architectural structure, we realized that the connection between architecture and anthropology was now perhaps manifesting itself in cultural studies and politics: Black Lives Matter and the confrontation of the Trump Government of the time threw up new questions concerning the control of space that architectural education has to confront. Such confrontation, involving terms and concepts proper to Pacific ontologies and epistemologies – together with their corresponding counterparts in the very Western knowledge traditions that colluded with colonization and exploitation – is difficult but urgent. Albert L. Refiti is a research leader in the field of Pacific spatial and architectural environment with an extensive research and publication experience in the area, supported by his teaching and lecturing in the last



Conclusion  191

fifteen years. His current research is on Pacific concepts of space – how they are formulated and enacted – the aim which is to find out how this understanding might play a role in rethinking the ways that Pacific people can create new modes of working and new notions of place and citizenship in the diaspora towards a Pacific cosmopolitic. A.-Chr. Engels-Schwarzpaul is Professor in Spatial Design and Postgraduate Studies at Auckland University of Technology – Te Wānanga Aronui o Tāmaki Makau Rau, in Auckland, Aotearoa/New Zealand. She was Executive Editor of  Interstices: Journal of Architecture and Related Arts  from 2005 to 2012  and edited approximately eight special issues for  Interstices,  ACCESS and Knowledge Cultures. She also edited  Of Other Thoughts: Non-traditional Ways to the Doctorate: A Guidebook for Candidates and Supervisors (2013, Sense Publishers, with M.A. Peters) and edited and translated  Atmospheric Architectures: The Aesthetics of Felt Spaces (2017, Bloomsbury, author Gernot Böhme).

Notes   1. As Lana Lopesi (2021) observes, ‘Moana and non-Moana researchers working in areas such as Guåhan, Solomon Islands and Micronesia did not feel an affinity to the term [Moana], and even felt excluded by those who use it. Similarly, participants in a talanoa hosted by Vā Moana / Pacific Spaces in June 2019, titled Salty Blood and Tears: Ocean stories, could not find a common ground and a connection to the term Moana’. While one participant felt that the term ‘did not reflect her or her language groups’, Vincente Diaz considers it problematic ‘when Ocean as Moana conflates one part of the Pacific for the entirety, and loses specificity of seas’, since not ‘much of the Pacific’ deals ‘with the Ocean as do peoples from Polynesia and Micronesia’ (Diaz 2019: 35).   2. See https://www.aucklandmuseum.com/your-museum/about/five-year-plan.   3. See https://www.mollerarchitects.com/hihiaua-cultural-centre.   4. Provided that the experts are self-reflexively aware of its function and the implication of its projects and work in close collaboration with the communities they seek to understand, rather than using them as informants (e.g. Memmott and Davidson 2008).   5. The Ecological Design Collective is a ‘community for radical ecological imagination and collaborative practice’, see https://ecodesigncollective.org.   6. This conceptual aspect of relationality appears to have moved to the background in current vernacular use of wā in Māori, which refers primarily to temporal aspects. The relational ones now seem to be articulated through terms such as whakapapa (to place in layers, genealogy) or whanaungatanga (relationship, sense of family connection).

192  Albert L. Refiti and A.-Chr. Engels-Schwarzpaul



  7. The posts, for example, do not sit in the earth but on the ceiling of an underground carpark; the ridge beam was replaced by a ventilation shaft; the fau (curved beams) are only decorative; and the ornamental lalava (lashing patterns) cover steel plates and bolted joints.   8. See MAHI TAAPOI Sustainable Tourism Development, http://www.mahitaapoi.co.nz/Kupe_Waka_Centre_files/Kupe%20Waka%20Centre%20 Project%20Outline%20August%202005.pdf.

References Arendt, Hannah. 1998. The Human Condition (2nd edn). Chicago: University of Chicago Press. Bedford, Richard. 2001. ‘Transformation of Domestic Environments: An Ignored Dimension of Transnational Communities’, in Richard Bedford, Robyn Longhurst and Yvonne Underhill-Sem (eds), Flowers, Fale, Fanua and Fa’a Polynesia. Wollongong: University of Wollongong. Centre for Asia Pacific social transformation studies. Asia Pacific migration research network (APMRN), published with UNESCO, pp. 1–5. Brown, Peter. 2002. ‘Review of Ethnologie et Architecture: Le Centre Culturel Tjibaou, une Realisation de Renzo Piano, by Alban Bensa’, The Contemporary Pacific 14(1): 281–84. Buchli, Victor. 2013. An Anthropology of Architecture. London: Bloomsbury. Cheng, Jennifer S. 2016. Other Ways of Seeing: The Poetics and Politics of Refraction. Retrieved 24 March 2021 from https://jacket2.org/commentary/ poetics-and-politics-refraction. Christiansen, Lene Bull, Lise Paulsen Galal and Kirsten Hvenegård-Lassen. 2017. ‘Organised Cultural Encounters: Interculturality and Transformative Practices’, Journal of Intercultural Studies 38(6): 599–605. Diaz, Vincente M. 2019. ‘Oceania in the Plains: The Politics and Analytics of Transindigenous Resurgence in Chuukese Voyaging of Dakota Lands, Waters, and Skies in Miní Sóta Makhóčhe’, Pacific Studies 42(1/2). Eisenman, Peter. 1984. ‘The End of the Classical: The End of the Beginning, the End of the End’, Perspecta 21: 155–73. Engels-Schwarzpaul, A-Chr. 2019. ‘Atmospheric Thresholds and the Production of Cross-Cultural Spaces’, in Sandra Karina Löschke (ed.), Non-Standard Architectural Productions: Between Aesthetic Experience and Social Action. New York: Routledge. Engels-Schwarzpaul, A.-Chr. and Carl Mika. 2019. ‘Silences Generating Space’, in Mark Dorrian and Christos Kakalis (eds), The Place of Silence: Architecture / Media / Philosophy. Milton Park, Abingdon: Bloomsbury. Hau‘ofa, Epeli. 1994. ‘Our Sea of Islands’. Contemporary Pacific 6(1).



Conclusion  193

Ingold, Tim. 2013. Making: Anthropology, Archaeology, Art and Architecture. Milton Park, Abingdon: Routledge. Lehman, Fred K. and David J. Herdrich. 2002. ‘On the Relevance of Point Fields for Spatiality in Oceania’, in Giovanni Bennardo (ed.), Representing Space in Oceania: Culture in Language and Mind. Canberra: The Australian National University, pp. 179–97. Lopesi, Lana. 2020. ‘Quiet Gestures, Gift Exchange and Public Formations: The Work of D.A.N.C.E. Art Club and Public Share’, in Cameron Cartier and Leon Tan (eds), Routledge Companion to Art in the Public Realm. New York: Routledge. . 2021. ‘Moana Cosmopolitan Imaginaries: Toward a New Theory of Moana Art’, PhD dissertation. Auckland: Auckland University of Technology. Losche, Diane. 2003. ‘Cultural Forests and Their Objects in New Caledonia, the Forest on Lifou’, Australian and New Zealand Journal of Art 4(1): 77–91. Lythberg, Billie. 2020. ‘Mediating Encounters and Economies’, in A. Bunbury (ed.), Pride of Place: Exploring the Grimwade Collection. Melbourne: Mlegunyah Press, pp. 58–61. MacPherson, Cluny. 1997. ‘A Samoan Solution to the Limitations of Urban Housing in New Zealand’, in Jan Rensel and Margaret Rodman (eds), Home in the Islands: Housing and Social Change in the Pacific. Honolulu: University of Hawai‘i Press, pp. 151–74). Mahina, ‘Okusitino. 2019. ‘Taha he Kehe – Unity in Diversity’, in James Bhagwan, Elise Huffer, France C. Koya-Vaka’uta and Aisake Casimira (eds), From the Deep: Pasifki Voices for a New Story. Suva: Pacific Theological College. Memmott, Paul and James Davidson. 2008. ‘Indigenous Culture and Architecture in the South Pacific Region’, Fabrications: The Journal of the Society of Architectural Historians, Australia and New Zealand 18(1): 74–117. Rapoport, Amos. 1991. House Form and Culture. Englewood Cliffs, NJ: Prentice-Hall. Refiti, Albert L., A.-Chr Engels-Schwarzpaul, Lana Lopesi, Billie Lythberg, Layne Waerea and Valance Smith. 2021. ‘Vā at the Time of COVID-19: When an Aspect of Research Unexpectedly Turns into Lived Experience and Practice’, Journal of New Zealand & Pacific Studies 9(1, New Scholarship in New Zealand and Pacific Studies). Refiti, Albert L, Rau Hoskins and A.-Chr Engels-Schwarzpaul. 2021. ‘Indigenous Architecture and the Politics of Resistance: Waipapa Marae and the Fale Pasifika at The University of Auckland’, in Nikolina Bobic and Farzaneh Haghighi (eds), The Routledge Handbook of Architecture, Urban Space and Politics (Vol. 2). London: Routledge. Ricoeur, Paul. 1978. ‘The Metaphorical Process as Cognition, Imagination, and Feeling’, Critical Inquiry 5(1): 143–59.

194  Albert L. Refiti and A.-Chr. Engels-Schwarzpaul



Rudofsky, Bernard. 1987. Architecture Without Architects: A Short Introduction to Non-Pedigreed Architecture. Albuquerque, NM: UNM Press. Rykwert, Joseph. 1988. The Idea of a Town: The Anthropology of Urban Form in Rome, Italy and the Ancient World. Cambridge, MA: MIT Press. Sarvimäki, Marja. 2019. ‘ANTi-History in Design Research’, Symposium Conducted at the Meeting of the ARCC International Conference - The Future of Praxis: Applied Research as a Bridge Between Theory and Practice, Toronto, 29 May – 1 June 2019. Toronto: Ryerson University. Smith, Linda Tuhiwai. 2012. Decolonizing Methodologies: Research and Indigenous Peoples (2nd edn). London: Zed Books Ltd. Tapsell, Paul. 2002. ‘Marae and Tribal Identity in Urban Aotearoa/New Zealand’, Pacific Studies 25(1): 141–71. Van der Ryn, Fepulea‘i Micah. 2012. ‘The Difference Walls Make’: Cultural Dynamics and Implications of Change in Sāmoan Architectural Traditions and Socio-Spatial Practices (1940–2006)’, PhD dissertation. Auckland: University of Auckland. Viveiros de Castro, Eduardo. 2012. ‘Cosmological Perspectivism in Amazonia and Elsewhere’, HAU: Masterclass Series 1: 45–168. Walker, Paul. 2018. ‘Architecture of the Contact Zone: Four Post-colonial Museums’, in Elizabeth Grant, Kelly Greenop and Albert L. Refiti (eds), The Handbook of Contemporary Indigenous Architecture. Singapore: Springer, pp. 927–53. Wendt, Albert. 1996. ‘Tatauing the Post-Colonial Body’, New Zealand Electronic Poetry Centre. Retrieved 23 April 2022 from http://www.nzepc.auckland. ac.nz/authors/wendt/tatauing.asp. Yaneva, Albena. 2009. Made by the Office for Metropolitan Architecture: An Ethnography of Design. [Unknown]: 010 Publishers.

Glossary ♦l♦

Māori ahuatanga Māori — Māori characteristics, traditions ariki — paramount chief aro — to pay attention to, to face toward something aroha — to love, feel pity, feel concern for ātea — courtyard, public forum, open area in front of wharenui hā — to breathe, taste hāngī — earth oven, food cooked in an earth oven hapū — subtribe hīkoi — to step, stride, march hoa — friend, companion, partner hoko — to buy, sell, barter hunga tiaki — group of guardians ihi — essential force iwi — extended kinship group, tribe kai — to eat, consume; food, meal kaitiaki — trustee, minder, custodian, guardian kaitiakitanga — guardianship, stewardship karanga — to call, call out; formal call, ceremonial call of welcome to visitors kaupapa — topic, policy, subject, programme kaupapa Māori — Māori approach, Māori topic, Māori customary practice kō — to locate, to dig kōhanga reo — Māori language preschool; literally, the language nest kōrero — to tell, say, speak, talk kotahitanga — unity, togetherness, solidarity, collective action kōwhaiwhai — painted scroll ornamentation mana — prestige, authority, power, influence manaakitanga — hospitality, kindness, generosity manaia — stylized figure used in carving Māoritanga — Māori culture, Māori practices and beliefs, Māori way of life marae — to be generous, hospitable; the complex of buildings surrounding the marae ātea, ceremonial/community space

196  Glossary

marae ātea — open area in front of the wharenui mātauranga Māori — Māori knowledge mauri — life principle, life force, vital essence noa — ordinary, unrestricted pā — fortified village pae — transitional zone of demarcation and negotiation pānui — public notice, announcement pātaka — storehouse raised upon posts pou — post, upright, support, pole, pillar poupou — wall-pillars, post, pole pou tokomanawa — centre pole supporting the ridge pole of a meeting house pou tuarongo — post supporting the ridge pole in the back wall of a meeting house pōwhiri — to welcome, invite, welcome ceremony on a marae pūkenga — specialist, expert pungawerewere — spider pūrākau — myth, ancient legend, intergenerational stories rangatira — chief; chiefly, noble, high ranking taonga — treasure, prized possession tapu — sacred, prohibited, restricted Te Ao Hou — the new world Te Ao Māori — the Māori world Te Ao Pākehā — the settler world Te Ao Tawhito — the old world te reo — the language, often used to refer to Māori language tikanga — correct procedure, custom, protocol tiki — carved figure in the abstract form of a human tino rangatiratanga — self-determination, sovereignty, autonomy tohunga — skilled person, chosen expert, priest, healer tohungatanga — expertise, competence, proficiency tuku — to present, offer, gift tuku iho — handed down tukutuku — ornamental lattice work tūrangawaewae — standing, place where one has the right to stand utu — to repay, pay, respond, avenge wā — time, season, period of time waiata koroua — traditional chant wairua — spirit, soul wairuatanga — spirituality waka taua — war canoe wānanga — to meet and discuss, deliberate, consider wehi — to be awesome; response of awe in reaction to ihi





Glossary  197

whaikorero — oratory, oration, formal speech-making whakairo — to carve; carved, carving tohunga whakairo — master carver whakairo rākau — wood carving whakapapa — genealogy, genealogical table, lineage, descent whānau — extended family, family group whanaungatanga — relationship, kinship, sense of family connection whakawhanaungatanga — process of establishing relationships, relating well to others whao haehae — a chisel for v-shaped lines whare — house, building, residence whare kai — dining hall whare manaaki — house to care for guests wharenui — meeting house, main building of a marae wharepaku — toilet, lavatory whare tapere — house of entertainment, theatre whare tīpuna/tupuna — ancestral house whare whakairo — carved house, meeting house whatu — stone whenua — land

Marshallese aelon kaal — new islands aelon kein ad — our islands alap — land owner and elder of the family bwebwenato — chatting and storytelling enra — shared resources, basket of food erak — funerary process jake jebol eo — provide life for others jebta — traditional Marshallese dance kamems — first birthdays omimono — handicrafts weto — family land parcel

Samoan afolau — see fale ‘aiga — family, kin, relatives, lineage Alofa — love alofisā — ceremonial gathering ali‘i — high ranking chief

198  Glossary

amopou — wall plate, longitudinal log resting on mainposts of house ‘aso — battens, thatch rafter ‘aso o le totonu — centremost spar in the roof of the faletele ‘aso tau — main spars in the roof of a faletele ‘aso vao — thatch rafters in the roof of the faletele atua — god/s aualuma — semi formal association of unmarried women ‘au‘au — ridge beam, ridgepole ‘ava — kava, beverage made from the kava root fa‘afetai — thanks fa‘alavelave — that which interferes with normal life and calls for special activity fa‘amaseiau — defloration ceremony fa’a Sāmoa — Samoan way fa‘ase‘e — skewed fa‘ase’e ‘i lalo o le amopou — skew below the straight top plate fa‘atautau — hang person or object fale — house faleafolau — larger meeting house with a longer itu (middle section) faleapa — open fale fale fa’aivi’ivi — a style of fale lacking the structural crossbeams falekuka — cook house faleo‘o — ordinary dwelling house falepālagi — European house falesāmoa — Samoan house faletalimalo — guesthouse faletele — roundhouse suited for visitors and meetings fale‘ula — a ceremonial building reserved for chiefs’ meetings and looked after by the taupou fatu-le-ulu-‘aso — top-end lashing of the ‘aso’ fatuga — battens Ua vaea i ulu fatuga — divided at the fatuga fau — make, construct faulalo — curved top plate fau tali‘aso — ‘aso receiver fautū — arched purlins fululupe — translates to doves tail and refers to a particular fale building method fono — meeting, council gafa — genealogy itu — middle section of a fale komiti — committee la‘au matua — underpurlins lagi — heaven





Glossary  199

lavelave — tangled luma — front ma — pure, clear; to be all destroyed mafua‘aga — origin malae — open space in the middle of the village mālie — be amusing, funny; gentle, slowly malu — shelter mamanu — be decorated with a pattern of design mana — supernatural power manaia — titled chief ’s son manu — animal matagofie — beautiful, lovely matai — chiefs in a Samoan village moamoa — end piece of the ridge pole mutiagiagi — deceitful nature, false appearance noa — indicating that something is of no importance, worthless, easy nu‘u — village o le pola motu i tua — the torn blinds at the back of the house pālagi — European, European ways poulalo — lashings poutū — central post si‘ufau — final piece of the faulalo soa — friend, companion, pair sumu — lashing ornament, fish tai — sea tala — unfold, undo, open out; round end sections of a fale talava — highest main purlins tamasā — birth of the first child taotao — hold down layer tapa-making — making of barkcloth by pounding the bark of the paper mulberry plant with a wooden mallet tapu — sacred, set apart, restricted tasi ‘ae afe — one, but a thousand (proverb) taulaga — for Samoans, both anchorage and sacrificial offering taupou — ceremonial maiden tausi — take care of, maintain, keep in order te‘e — propping layer ti‘eti‘e — ride freely togafiti — trickery toloa — duck tua — back tua‘oi — neighbour, boundary

200  Glossary

tufuga — craftsman, expert, specialist tufuga-faufale — expert house builder matai tufuga — chief builder tulāfale — orator, talking chief tuli — chase, pack off; joint of the body tulutulu‘apa — timber-framed eaves to enlarge the overhang ‘ula — colour red, red feathers, red beaded ceremonial necklace ulu‘aso — the fixing method of the ‘aso at their upper ends uta — bush side vā — relational space vā moana — Pacific space teu le vā — to tend to the relational space vā fealoaloa‘i — social relationships, relations based on mutual respect

Hawaiian ‘aha — cord ‘aha hele honua — cord for binding the land and measuring the heiau ahu hō‘ola — an altar of healing akua — god/s āina — land anu‘u — oracle tower for praying and receiving guidance from the divine ‘auamo — carrying stick ‘aumākua — ancestral gods ‘awa — kava hale — house, dwelling hale mana — thatched houses for storing the temple god images hale o Papa — heiau for female rituals hale pahu — building for storing the sacred hale umu — building for the underground oven to bake sacred foods hale waiea — building for storing god images heiau — place of worship heiau hō‘ola — sacred place for treating and healing the sick ko‘a — heiau for offerings to ensure good fishing luakini heiau — sacrificial temple mua — heiau for male ceremonies waihau — heiau structures for offerings to increase food crops kāli‘i — rite kalo — taro kapu — tapu kāula — prophet Ke Ala Ma‘awe‘ula a Kanaloa — The Red Track of Kanaloa





Glossary  201

Ke Ala Polohiwa a Kanaloa — The Black Shining Path of Kanaloa Ke Ala ‘Ula ‘a Kāne — The Bright Red Path of Kāne kīhei — tapa cloth shawl ki‘i — images of deities stand steadfast like guardians or sentinels kilo — possibly a stargazer, astronomer, or seer kinolau — body form kīpapa — stone pavement for placing offerings kō — sugarcane kuleana — responsibility lā‘au lapa‘au — Hawaiian herbal healers lana — the lowest tier for offerings lele — altar lua — pits where remains of the victims were thrown luakini — sacrificial temple luapa‘u — refuse pit mahi‘ai — a planter, cultivator of food makahiki — an ancient festival for the Hawaiian new year Makali‘i — Pleiades Mākau a Maui — Fishhook of Maui, Scorpio malo — male tapa loincloth mamao — highest most sacred tier for the high priests and paramount chiefs manu — bird māpele — agricultural heiau momoa — small projection at the stern mo‘olelo — deep history mu‘a — in front nu‘u — the second tier for high priests and attendants ohana — family ‘ō‘ō — digging stick pahu — drums used in temple rituals papa-kuhikuhipu‘uone — kahuna or priest responsible for building and locating temples palaoa — a sperm whale piko — navel pōhaku — sacred healing stones ulua — giant trevally unu — horseshoe or ‘open mouth’ wā — time wa‘a — boat, canoe Wākea — Orion

202  Glossary

Tahitian marae — temple Marae Vaeara‘i — Temple of Heavenly Footprint Ra‘iātea — Tahitian

Tongan fala — pandanus mat fatongia — duty tā — time tā-vā — time-space talanoa — dialogue tāno‘a — kava bowl taula — priest tō — sugarcane tufunga fonua — principal architect of the Tongan socio-cultural obligations tufunga langafale — house builder vā — space vaka — boat, canoe



Index ♦l♦

Note: H=Hawaiian, M=Māori, RMI=Republic Marshall Island, S=Samoan, T=Tongan A aelon kaal (RMI: new islands), 130, 134, 136, 139, 142 ahuatanga Māori (M: Māori characteristics, traditions), 157 ali‘i (S: high chief), 39, 45, 62–63, 101, 104. See also ariki Allen, Chadwick, 3, 123 alofa (S: love). See also lē alofa, 6, 42–46 alofisā (S: sacred meeting circle), 45, 105, 109 altar, 97, 99–105, 108, 117 ancestors, 3, 15, 76, 81, 84, 88, 106–107, 114, 123–124, 136, 150, 169 ancestral geography, 7, 117, 124 anthropology, 2, 11–23, 116, 118, 184–185, 187, 190 anthropological lens, 167, 186 architectural anthropology, 6, 11–23 ethnography, 6, 54–74, 75–95 appropriation/appropriate, 16, 90–91, 162, 168 architecture, 11–23, 134, 184–186, 189, 190 classical European buildings, 96–112 Māori architecture, 11–23, 89, 148–158, 159–183 Pacific architecture, 96–112, 114 Samoan architecture, 24–36, 43–45, 54–74, 159–183

Zombie architecture, 96–112 ariki (H & M: paramount chief), 104, 149. See also ali‘i artefact/artefacts, 37, 43, 45, 84, 107, 148, 156, 162, 171, 173–174 astronomy, 118 atu (M & S: away from speaker), 41 authenticity, 17, 66, 86, 154, 173, 176 ‘ava ceremony (S: the imbibing of kava at the meetings of matai), 40, 43, 65–68, 105, 109, 123–124 ‘ava/‘awa/kava, 40, 43, 63–68, 105, 109, 120, 123–124 B Bennardo, Giovanni, 41–42 biopolitics, 91, 93 bloodletting, 97 Bourdieu, Pierre, 37–38 Burkert, Walter, 98–99, 104 bwebwenato (RMI: chatting and storytelling), 134–136, 142 C carving, 13–16, 35, 75–95, 115, 152, 154, 159, 164 Casey, Edward, 57 Christian/Christianity, 17, 32, 42, 96, 133 Clandon House, 150, 159 Colonization, 77, 130, 148, 156, 165, 172, 178 conservation, 13, 18–19, 117, 152, 168, 173 construction, 18, 24–36, 43, 45, 60, 108, 117, 135, 142, 162 cosmology, 44–45, 98, 100, 107, 190 cosmos/cosmic, 44–45, 100, 103, 117

204  Index

D Diaz, Vicente, 129, 142, 144 discipline/disciplinarity, 18–21, 37–38, 92, 177, 184 E enclaves, 130–131, 134 ethnology, 11–13, 167 F fale (S: house), 37–53, 57–67, 98, 115, 159–161, 163–164, 167, 175, 186, 189. See also whare, hale faleafolau (S: long house), 28, 43–46, 70 faleapa (S: open fale), 58, 70 fale fa‘aivi‘ivi (S: a style of fale lacking the structural crossbeams), 70 falekuka (S: cook house), 58, 70 faleo‘o (S: ordinary dwelling house), 34, 58, 70 falepālagi (S: European house), 58, 70 falesāmoa (S: Samoan house), 58 faletalimalo (S: guesthouse), 30, 32, 34, 70 faletele (S: round house suited for visitors), 28–29, 32, 43–46, 70, 101–103, 167, 176 fale‘ula (S: ceremonial building reserved for chiefs’ meetings and looked after by the taupou) 101–102 force field, 41–42 point field, 41–42 Field Museum (Chicago), 153, 175, 177 fono (S: meeting, council), 26, 35, 59–68, 105, 109 G gafa (S: genealogy), 25, 43. See also genealogy genealogy/genealogies/genealogical, 15, 25, 45, 79, 123, 134, 149, 165, 166, 177. See also gafa, whakapapa

♦ H hale (H: house, dwelling), 115, 116 hale mana (H: thatched houses for storing the temple god images), 115 hale o Papa (H: heiau for female rituals), 116 hale pahu (H: building for storing the sacred), 115 hale umu (H: building for the underground oven to bake sacred foods), 115 hale waiea (H: building for storing god images), 115 Hawaiian Tempo-Spatial Structure, 113–128 heiau (H: place of worship), 97, 104–105, 109, 113–128 heiau hō‘ola (H: sacred place for treating and healing the sick), 116–117 ko‘a (H: heiau for offerings to ensure good fishing), 115 luakini heiau (H: sacrificial temple), 97, 104–105, 115, 118 Maunawila Heiau, 113–128 Mua (H: heiau for male ceremonies), 116, 123 waihau (H: heiau structures for offerings to increase food crops), 105, 115 Herdrich, David, 41–42 Hersey, George, 97, 98, 100–101 Hinemihi o te Ao Tāwhito (she/her, Hinemihi of the Old World), 148, 150, 167, 168, 172–175 hoa (M: friend, companion, pair), 3, 4. See also soa House of Knots, 75–95 I iconicity, 59, 81, 86, 97, 107, 121, 173, 189-90 identity, 14, 19–20, 46, 75, 134, 141–142, 144, 148, 156, 172, 173 Indigenous/Indigeneity, 17–21, 77, 81–84, 89, 117, 123–124, 129, 134, 144–145, 149, 152, 172

♦ Indigenous knowledge, 11–21, 89, 92, 189 trans-Indigenous, 123–124 Ingold, Tim, 185 itu (S: middle section of the roof), 24–26, 43 K Ka‘ili, Tēvita, 3, 4 kaitiaki (M: guardian), 149, 170 kaitiakitanga (M: guardianship), 171, 173 kaupapa (M: initiative, agenda, issue, values), 79, 149, 168 kaupapa Māori (Māori research principles), 14, 18–21, 92, 153 kō-design, 148, 157 Kōhanga Reo (Māori language preschool; literally, the language nest), 150, 159 kotahitanga (unity, togetherness, solidarity, collective action), 150 L lagi (S: heaven), 44, 103 lē alofa (S: devoid of alofa), 42–43. See also alofa Lefebvre, Henri, 54–61, 68 theories of space, 55–57 Lehman, F.K., 41–42 Linden Museum (Stuttgart), 153, 175 M mafua‘aga (S: origin), 26–28 mai (M & S: towards speaker), 41 malae (S: open space at the village centre), 45, 59–67, 101, 104–105, 107 mana (S: power; M: prestige, authority, control, power, influence, status, spiritual power), 17, 42–46, 80, 83–84, 102, 105, 114–115, 117, 190 manatu lelei (S: good thoughts), 42 Māori (tangata whenua, Indigenous people of Aotearoa New Zealand), 11–21, 75–95, 148–158, 159–183, 186–189 marae (M: ceremonial space), 15–16, 81, 85, 107, 117–118, 149, 156, 159, 162, 175, 188

Index  205

marae ātea (M: open space in front of the wharenui), 159, 177 matai (S: chief), 43–45, 62–64, 66, 163 mātauranga (M: knowledge), 13, 20, 149, 154 migration, 13, 38–39, 113, 129–138, 190 Moana, 2–7, 113–128 mo‘olelo (H: deep history), 113, 115, 123 mounds, 96, 103–105, 107 Museum für Völkerkunde (Hamburg), 153, 168–169, 170, 173 N Ngā Kohinga Whakairo o Hinemihi, 153 Ngata, Āpirana, 11, 13 Ngāti Hinemihi, 149–150, 152, 153, 157, 162 Ngāti Rānana, 150, 153, 159, 162, 174 Neich, Roger, 17, 166 noa (M, S & T: ordinary, unrestricted), 105, 167, 175, 177, 189. See also tapu O Onslow, William Hillier, 150 openness, 15, 24, 169 oppositive dualism, 64 P Pacific, 19. See also Moana Pākehā (M: non-Māori, particularly of European descent), 16, 90, 150 place-making, 130–134, 136, 138, 140–141 Rimajol place-making, 129–147 pou (M & S: pole), 27–34, 43–44, 76–77, 102, 106 poutū (S: main pole), 43, 103 power, 45–46, 56, 61, 84, 88, 114, 123, 132, 141 production of space, 56–57 pūkenga (M: expert), 150 R rangatira (M: leader), 148, 149 Rauru, 19, 153, 160–163, 167–172 reanimation, 97

206  Index

relationality, 90, 130, 134, 185, 187, 191 relational space, 37–53, 56–57, 81–82, 129, 187–188 relationships, 15, 19, 39, 42, 56, 59–69, 80–81, 87, 103, 134, 136, 142, 143, 164–167, 169–173, 184–190 social relationality, 134, 160, 166, 167, 189 social relationships, 65–69, 132, 136, 188, 189 religion/religious, 42, 142 Republic of the Marshall Islands, 129–147. See also Rimajol Rimajol, 129–147. See also Republic of the Marshall Islands ritual, 15, 43–45, 66, 81, 96–100, 105, 108, 120 ritual attractor, 97–98, 108 roots and routes, 129 Ruatepūpuke, 153 S sacred, 45–46, 55, 79, 85, 88, 91, 98, 101–107, 113–124 sacred buildings, 96–97, 108, 109 sacrifice, 96–112, 118 Salmond, Anne, 15–17, 81, 90 Sā-Tagaloa (S: Tagaloa family), 26, 44. See also Tagaloa, Tagaloalagi settler/settlers, 12, 77, 80, 90, 106, 130, 145 Shore, Bradd, 42–43, 45 soa (S & T: friend, companion, pair), 3, 4. See also hoa space, 97, 100, 105, 129–144, 184–185, 187–190 Pacific space, 35, 108, 123–124, 129–144, 184–185, 187–190 relational space, 37–53, 56–57, 81–82, 129, 187, 188 representational spaces, 55–58, 61–64 spatial practice, 55–58, 60–63, 67–68 spatiality, 41–42, 130 spirituality, 17, 76, 150 Strathern, Marilyn, 90

♦ T tā (T: time, beat), 117, 123, 124 Tagaloa, 26, 44–45, 71, 101, 109, 121 tala (S: round end of the house), 24–36, 43, 45 tangata whenua (M: people of the land, locals, Indigenous people), 189, 190 tapu (M, S & T: scared, set apart, restricted), 42–43, 45–46, 77, 97, 101, 104–105, 114, 189. See also noa Taroi, Wero, 149 Te Ao Māori (M: the Māori world), 80, 90 Te Maru o Hinemihi, 150, 162, 174 Te Tiriti o Waitangi (M: Treaty of Waitangi), 13, 148, 174 Te Wharepuni a Mauī, 153, 175 tectonics/tectonic, 103 indigenous tectonic traditions, 14, 24–36, 103 temple, 97–99, 101, 103, 105–106, 109, 114, 115, 117, 118. See also heiau megalithic temple, 114 Tengan, Ty, 114, 116, 123 tikanga Māori (M: Māori custom, protocol), 20, 76–77, 79–80, 175 tino rangatiratanga (M: self-determination), 16, 20 tohunga (M: skilled person, expert), 14, 81, 162. See also tufuga, tufunga Tomlinson, Matt, 42 threshold, 2 transtransboundary, 131, 136, 143 transdisciplinarity/ transdisciplinary, 3, 5, 55 transindigeneity, 123–124 translation, 4–5, 40, 161, 164–168, 170–173 transnationalism, 130–133, 142–144 transplantation, 129 travelling houses, 159–183 tufuga (S: expert, specialist), 27, 34–35, 43, 98, 163, 164. See also tohunga, tufunga tufuga-faufale (S: expert builder), 26–27, 34–35, 43, 98, 164 tufunga (T: expert, specialist), 98, 122. See also tohunga, tufuga

Index  207

♦ Tūhourangi, 150, 152, 174 tulāfale (S: talking chief), 39, 62–63, 66, 71–72 U UK Arts Council, 152 UK National Trust, Heritage New Zealand Pouhere Taonga (NZHPT), 152 United States, 129–147 University College of London’s Institute of Archaeology, 152 utu (M: reciprocity), 166 V vā (S: relational space), 37–53, 54–74, 117, 124, 129, 184–190 tā-vā (T: time-space), 117, 123, 124 vā-energy, 45 vā fealoaloa‘i (S: relational space and social relationships), 39–46, 166–167 vā-field, 37, 42–43, 46 vā-forces, 42, 45–46 vā relations, 38–46, 65 vā tapua‘i (S: sacred space), 39–46 Valeri, Valerio, 102–105 vector (see vā-forces), 41–42, 59, 63 W wā (H: time), 187, 188 Waitere, Tene, 150, 162

wānanga (M: discussion, learning, expert, knowledge), 149 welfare, 76–81, 84–85, 90–91 Wendt, Albert, 38–39, 109, 187 whakairo (carving), 14–19, 79, 156. See also carving, whare whakairo whakapapa (M: genealogy, lineage, layering), 15, 90, 149, 150, 154, 165, 166, 168. See also gafa, genealogy Whakarewarewa, 153 whakawhanaungatanga (M: relationship building, inclusive relations), 89, 149 whare (M: house), 98, 115, 149, 150, 154, 156, 161, 162, 168, 174, 175 whare manaaki (M: house to care for guests), 174 whare tapere (M: house of entertainment), 150 whare tīpuna/tupuna (M: ancestral house), 106, 161, 162, 168 whare whakairo (M: carved meeting house), 12–16, 19, 148, 149, 150, 156, 161, 173, 176 wharenui (M: big house, meeting house), 81, 156, 160, 162, 174, 186 whenua (M: land), 80, 149 woodcarving, 75–76, 79, 150, 152–153 Z zombification, 97–98, 106