H awaiki, A ncestral Polynesia.

A n E ssay in H istorical A nthropology by Patrick W inton

In recent decades the historical and com parative research in the languages o f O cean­
ia has expanded considerably linguistics and its m ain centers are A ustralia, N ew Z ealand
and A m erica (esp. H aw aii). Its qualitative advance m ay be attributed at least partly to
the m ultidisciplinary approach o f its protagonists w ho do not view anthropological and
archaeological data as irrelevant for historical and com parative research in the field o f
linguistics. W ithin this m ethodological fram ew ork linguistics m ay be said not only to d e­
scribe, but also to explain. This turn tow ard cognitive issues and m ultidisciplinarity m ay
be observed in m any studies and books dealing w ith the Pacific. O ne o f them is H a ­
waiki, A ncestral P olynesia (An E ssay in H istorical Anthropology) prepared by P atrick
V inton K irch and R oger C. G reen and published in 2001 in Cam bridge U niversity Press.
In the Preface the authors underline the universality o f their anthropological educa­
tion, interests, and research in quite a few P olynesian societies and their inclination to
a com prehensive investigation o f historical processes in Polynesia. O f course, attem pts
to restrict history to the era reflected in w ritten docum ents or inscriptions w ere rejected
a long tim e ago. The occasional tension betw een linguistic data obtained through ety­
m ology and sem antic m ethods on the one hand and archaeological data on the other is
som etim es undeniable; it is not easy to judge the ethnic and/or linguistic affiliation o f an
ancient society from archaeological data in a territory open to invasions and m igrations.
From this point o f view Polynesia is in an advantageous position and linguistic data m ay
help us to reconstruct extralinguistic phenom ena.
The term Hawaiki ranks am ong those Polynesian words that are relatively well known
outside the fairly narrow circle o f Polynesian scholars and maybe that is why the authors
have chosen it for the title o f their joint Work. In fact, they have launched a quest for the dis­
covery o f what is shared by all Polynesians - not only by those in the East for whom Hawaiki
probably was in the West Polynesia as indicated by the westernmost toponym Savai'i in the
archipelago o f Samoa. Their publication is m eant as a summary o f w hat we know o f the an­
cestral Polynesian phase or world irrespective o f its geographical location and name.
Their m ethod is derived from com parative historical linguistics endeavouring to re­
construct the proto-language and to explain changes that have occurred after the original
unity disintegrated into a num ber o f daughter languages. The so-called p hylogenetic
m odel is relatively easy to apply to a fairly restricted set o f com m unities sharing - in ad­
dition to language - sim ilar physical, social and cultural features for w hich we m ay as­
sum e a com m on origin w hile the subsequent changes have resulted from the need o f ad­
aptation. Interference w ith other ethnic units m ay also be included in the adaptation, but
it has played no significant role in (at least East) Polynesia. A b rie f instructive character­
istics o f the phylogenetic m odel and its application to Polynesia is described on pp. 13-
16. K irch and G reen trace their m odel to A. K. R o m n ey 's paper published in 1957 being
aw are, how ever, that he had predecessors as early as the 19th century. In K irch and
G reen's H aw aiki linguistics is no t m erely a m odel o f com parison but the acquired lin­
guistic data (cognates, etc.) are o f relevance for their decision m aking.
In the introductory chapters the theoretical principles and m ethodological procedures
for correlating linguistic and archaeological evidence are explained. In Chapter 3, the argu­
ments supporting the definition o f Polynesia as a clear-cut phylogenetic unit are discussed.
The establishm ent o f the particular daughter com m unities is not view ed only in term s
o f their initial separation; subsequent contacts upon the parallel level (i.e. w ith other

96
geographically not too distant daughter com m unities) are considered probable, although
m uch less so contacts w ith the hom eland community. Such contacts m ay have been due
to chance voyages o f individuals or to m ore m assive events that deserve to be term ed
intrusions or invasions (p. 33). A nother issue o f interest in this respect is the w here­
abouts o f the hom eland o f the daughter languages. It is often linked to the greatest genet­
ic diversity o f a region. H ere a question arises w hether the considerable internal diversi­
ty w ithin the M arquesas is to be ascribed to a relatively long-term diversification (w hen
voyaging betw een the islands w ithin the archipelago was not too com m on) or if it m ay
be taken as an indication o f the fact that th e diversification o f the local dialects is due to
the role o f the M arquesas as an early center o f m igrations w ithin E ast Polynesia.
The plausibility o f results achieved by K irch and Green is increased by the requirement
that “the subdisciplines o f historical linguistics, archaeology, comparative ethnology, and bio­
logical anthropology independently contribute their data and assessments to the common ob­
jective o f historical reconstruction” (p. 42). This procedure is metaphorically labeled the tri­
angulation m ethod by the authors. Parallel application data from several (sub)disciplines to
the reconstruction o f past phases is know from other areas - for example from Indo-Europe­
an studies. In the field o f Austronesian studies repeated attempts have been made for exam ­
ple by A. Pawley, M. Ross, M. O sm ond and others to investigate semantic fields and to use
them for explaining the cultural history o f the communities concerned.
In fact Kirch and Green proceed to their goal o f reconstructing the Polynesian homeland
from two directions - from the present to the past (by comparing the m odem languages and
culture w ithin Polynesia) but also (chiefly thanks to archaeology and to com parison o f
Polynesia with other parts o f Oceania) they perceive this reconstmcted phase as a result o f
w hat had been taking place before. In other words, they proceed from the earlier past to the
more recent past (Chapter 3, pp. 53-91). In this chapter, die authors are weighing up the role
o f isolation bz distance and are inclined to reject too pessimistic an attitude (pp. 83-89).
In Part II titled R ediscovering H a w a iki although they are aw are that the aspect o f
constructing is inevitably present in such an endeavour. A nd the preference w as given to
rediscovering because their aim w as not to hide their intentions.
In the subsequent chapters K irch and G reen concentrate upon the analysis o f a series
o f sem antic fields relevant for the reconstruction o f the ancestral Polynesia, First o f them
is the physical environm ent (see p. 103) follow ed by subsistence, food preparation and
quisine, m aterial culture, social and p olitical organization, gods, ancestors, seasons, and
rituals astronom ical phenom ena, w eather, directionals (p. 104), etc. In each instance an
inventory o f reconstm cted P roto-P olynesian w ords is listed including their cgnates in the
m odem languages w ith glosses and sources.
The conclusions proposed by the authors are plausible and not surprising. Their at­
tem pt at a reconstruction o f H aw aiki is a kind o f extension o f ethnography into the past
w ithout being a simple projection o f the present into the past. D espite their generally small
size the Polynesian com munities and their organization are far from sim ple and we are re­
m inded that two m illennia ago the ancestral communitities were extrem ely tiny (p. 282). A t
the same time we should be aw are that there was a good deal o f variety in all respects (in­
cluding dialectal differentiation) and flexibility was one o f the vital presuppositions o f effi­
cient evolution and purposeful adaptation to new environmental conditions.
The publication is supplied w ith abundant notes (pp. 285-312), a glossary (pp. 313-
316), a bibliography (pp. 317-355), and tw o indices including Proto-P olynesian recon­
structions (pp. 356-375). K irch and G reen have no doubt added a m ost interesting w ork
to the fundam ental library o f Polynesian studies.

Viktor K rupa

97