Wealth in old Hawai‘i: good-year economics and the rise of pristine states

THOMAS S. DYE

T. S. Dye & Colleagues

ABSTRACT

The journals of Captain Cook and his crew contradict bad-year economic theories that posit that traditional Hawaiian farmers were

living at the margin. Recognising that pig herds were wealth-assets in old Hawai‘i, an alternative good-year economic theory is

developed that interprets the introduction of sweet potato and the development of the rain-fed agricultural systems in which it was

cultivated as processes in the creation and management of wealth. The wealth produced in this way was probably used, in part, to

promote marriage alliances among elite families. According to the good-year economic theory, ﬂuctuations in the products of the

rain-fed agricultural facilities introduced variability into the supply of wealth-assets, which complicated the maintenance of alliances

and were one cause of the wars that played a crucial role in the emergence of primary states in traditional Hawai‘i.

Keywords: Bayesian radiocarbon calibration, Hawai‘i, primary state, rain-fed agriculture, sweet potato, wealth.

RÉSUMÉ

Les journaux du « Captain » Cook et de son équipage viennent en contradiction des théories économiques « de mauvaise année », qui

placent les horticulteurs hawaiiens traditionnels en limite de productivité d’autosufﬁsance. Reconnaissant que les troupeaux de cochons

constituaient des produits de richesses dans l’ancien Hawai’i, une théorie économique alternative « de bonne année » est proposée,

interprétant l’introduction de la patate douce et la multiplication de systèmes agricoles pluviaux permettant sa production, comme des

processus de création et de gestion de richesses. La richesse ainsi produite était utilisée, en partie, pour la promotion d’alliances de

mariages entre les élites. Suivant la théorie économique « de bonne année », la ﬂuctuation dans les productions issues des modes

agricoles pluviaux engendrait des variations dans l’accès aux biens formant les richesses, compliquant le maintien d’alliances et

devenant une des causes des guerres qui jouèrent un rôle crucial dans l’émergence d’états primitifs au sein du Hawai’i traditionnel.

Mots Clés: Calibration radiocarbone Baysienne, Hawai’i, état primitif, agriculture pluviale, patate douce, richesses.

Correspondence: Thomas S. Dye, T. S. Dye & Colleagues, 735 Bishop Street, Honolulu, Hawaii 96813, USA. Email:

tsd@tsdye.com

INTRODUCTION

Were the Hawaiian maka‘a¯inana – the common people

–

living at the margin, frequently in danger of going hungry

during bad years? This is a question raised by two recent

book-length investigations of traditional Hawaiian society

(Hommon 2013; Kirch 2010a), each carried out in a

neo-evolutionary framework that stresses imbalances

between population and resources as a dominant locus of

social change. They posit a history of Hawai‘i where the

agricultural system yields small or no surpluses and

minimal subsistence requirements of farmers are in danger

of not being met, a region of the production function

model that can be labelled “bad year” (Figure 1). This

density-dependent, bad-year economics approach to

Hawaiian history has its basis in one of two

neo-evolutionary models proposed a half century ago when

social scientists began to synthesise information on

Polynesian societies with the goal of explaining the

differential development of social complexity in Polynesia.

The model that links social change to density-dependent

mechanisms had its genesis in the early work of Sahlins

(1958), while the other approach, more closely aligned

with structuralism, theorised density-independent

mechanisms of change situated in the pan-Polynesian

cultural theme of status rivalry (Goldman 1970).

Although the investigations of Kirch (2010a) and

Hommon (2013) both depend on bad-year economics, they

characterise bad years somewhat differently. Hommon

develops a “hard times hypothesis” (Hommon 2013: 241)

based on a description of social change on the Polynesian

outlier of Tikopia following a hurricane in 1952 that

destroyed crops and led to severe food shortages (Firth

1959). In the face of escalating thefts from gardens,

bs_bs_banner

Archaeology in Oceania, Vol. •• (2014): ••–••

DOI: 10.1002/arco.5034

Tikopia chiefs assumed authority to regulate access to

garden lands and imposed increasingly harsh punishments

on thieves. Hommon argues that, in a similar fashion,

“large-scale food stress” (Hommon 2013: 158) on Hawai‘i

Island and perhaps elsewhere provided the context for

Hawaiian chiefs to augment their power in ways that

transgressed sociopolitical norms and thus spurred the

development of social complexity. Kirch takes a different

tack, relying on Malthusian theory with a strong focus on

the relation of agricultural yield to population size

(Hommon 2013: 217 ff.; Kirch 2010a: 177 ff.; Kirch &

Zimmerer 2010; Kirch et al. 2012; Lee & Tuljapurkar

2008, 2010; Puleston & Tuljapurkar 2008; Tuljapurkar

et al. 2007) to argue that Hawaiian population growth

began to slow in the sixteenth century because production

was reaching practical limits, a situation that increasingly

diminished a chief’s prospects of laying claim to

surpluses.

In both accounts, a primary cause of bad years was the

variable yield of rain-fed agricultural systems (Hommon

1986, 2010, 2013; Kirch 1984, 1994, 2010a,b). This idea

is based on an analytical model that distinguishes

predictable and high-yield irrigated agricultural systems

from less predictable and poorer yielding rain-fed

agricultural systems (Barrau 1965). The analytical model

was historicised and humanised for Hawai‘i when Kirch

(1994: 251–68) associated the development of rain-fed

agricultural systems with the rise of a powerful line of

Hawai‘i Island ali‘i, or chiefs, leading to the great king

Kalani‘o¯pu‘u and his successor, Kamehameha, who

famously succeeded in uniting the Hawaiian Islands under

his rule in the late eighteenth century. In the

density-dependent theory, variable yields from rain-fed

agricultural systems on Hawai‘i and Maui Islands at times

dipped below the level needed to feed agricultural workers

and reduced the growth in tribute that could be collected

by chiefs, leading them to wage wars of conquest with

their more afﬂuent irrigated agricultural neighbours.

According to theory, the organisational skills gained

during warfare and later when managing conquered lands

initiated a dynamic that spurred the rise of social

complexity, culminating in the emergence of state

institutions (Hommon 1976, 1986, 2013; Kirch 2010a,b).

A density-independent approach was elaborated by

Earle (1978), who argued that development of agricultural

facilities would beneﬁt “both the elites and their dependent

population” (Earle 1978: 183) and thus that agricultural

intensiﬁcation “was an outcome of political competition

and not of population pressure” (Earle 1978: 183). This

paper attempts to build on the density-independent line of

reasoning by considering how social complexity might be

achieved during good years. It argues that the production

function model used by bad-year economists is too simple

for the Hawaiian situation, and that the broad category of

surplus production must distinguish between consumables,

which contribute to well-being and quality of life, and

wealth, which is “foundational to the power of the

dominant gender, social caste or class in every society”

(Bell 2004: 12).

In particular, cross-cultural characteristics of

wealth-assets (Bell 2004: 12–17) make it possible to

identify pig herds as a powerful wealth-asset in old

Hawai‘i, the only island group in Polynesia where pigs

were “numerous enough to be classed as a capital reserve

available for exchange” (Goldman 1970: 476). Because

“exchanges are the code through which status information

is communicated” (Goldman 1970: 496), development of

wealth-assets is tied to status rivalry. In this view, the

pursuit of status, in part through the exchange of pigs,

spurred the development of social complexity.

These two explanatory approaches – a

density-dependent approach based on bad-year economics,

and a density-independent approach based on good-year

economics – are not mutually exclusive. Hawaiians

certainly knew good years when vegetable foods were

plentiful and pig herds were healthy and growing, and bad

years when vegetable crops and pig herds produced less

than expected. Experiences of both kinds would have

shaped habits and activities, contributing to the production

and reproduction of the social forms encountered by

Europeans in the eighteenth century. From a theoretical

point of view, both approaches would seem to be required

for completeness. For the archaeologist, the choice of one

approach over the other, or the degree to which they each

contribute to an analysis, ought to be a practical matter

having to do with how well the approach structures

observations of archaeological materials and how

convincingly it relates those observations to a scientiﬁc

law or a historical explanation based on human intention

and motivation (Djindjian 2001).

A central argument of this paper is that the

density-dependent approach articulates poorly with the

archaeological record of traditional Hawai‘i.

Density-dependent interpretations rely on Hawaiian

Figure 1. A model of the relationship of population and

agricultural system output: T, agricultural production

function; V, minimal subsistence requirements to maintain

the working population; S, surplus production.

Source: Kirch (2010a: 194).

2 Wealth in old Hawai‘i

tradition and other synchronic sources for evidence of

what happened in old Hawai‘i (Hommon 2013: 212–16;

Kirch 2010a: 77–123); the problem of equiﬁnality leaves

these synchronic data open to a variety of plausible

interpretations.

This paper attempts to demonstrate that a

density-independent approach provides the structure

needed to interpret model-based chronologies recently

proposed for the development of agricultural facilities in

the rain-fed agricultural system of leeward Kohala district

on Hawai‘i Island, and for the construction of mostly

small, family-sized temples nearby.

Model-based chronology building using Bayesian

calibration of

C age determinations, which is able to

incorporate a wide range of chronological information,

produces precise results and yields age estimates for

events previously considered undatable, has the potential

to revolutionise archaeological interpretation (e.g. Whittle

et al. 2011). Here, it produces archaeological sequences of

change that are interpreted as the direct material remains

of wealth creation and management integral to the

development of social complexity in old Hawai‘i.

A MOST EXTRAORDINARY HOG ISLAND

Bad-year economics paints a dour picture of old Hawai‘i.

It interprets: (i) archaeological evidence as indicating a

sufﬁciently large population to put pressure on

agricultural production (Hommon 2013: 234–5; Kirch

2010a: 138–40, 216); (ii) the development of facilities to

intensify production in rain-fed agricultural systems as a

process of agricultural involution (Geertz 1963), where

additional labour inputs are rewarded with proportionally

smaller increases in yield (Hommon 2013: 233–4; Kirch

2010a: 147–8); (iii) measurably lower soil nutrient levels

within rain-fed agricultural soils as indications of

potential crop failure and/or declining yields (Hommon

2013: 232–3; Kirch 2010a: 149–50); (iv) Hawaiian

traditions as documents of famines that provoked social

unrest and conﬂict (Hommon 2013: 238; Kirch 2010a:

199–200); (v) ethnographic observations that Hawaiian

women worked in rain-fed agricultural ﬁelds on Maui

and Hawai‘i Islands, an unusual practice in Polynesia, as

evidence of labour shortage in the face of declining

yields (Hommon 2013: 61; Kirch 2010a: 139, 196, 216);

(vi) variable yields of rain-fed agricultural systems in

Hawai‘i as a cause of warfare that played a central role

in the development of social complexity (Hommon 1986,

2010, 2013; Kirch 1984, 1994, 2010a,b); (vii) the

relationship of ali‘i to maka‘a¯inana as a source of social

tension (Hommon 2013: 217–56; Kirch 2010a: 190–201);

(viii) the motives of chiefs in the Kamehameha line as

“hostile and expansionistic” (Kurashima & Kirch 2011:

3673); and (ix) archaeological evidence as indicating a

decline and abandonment of the rain-fed agricultural

ﬁelds soon after European contact (Kirch 2010a: 153)

and their rapid replacement in the landscape by “early

historic ranching enclosures, homesteads, and other

features of the postcontact era” (Ladefoged & Graves

2010: 89).

Given this, one might expect the initial accounts of

Hawaiian society to describe a parlous state of affairs

and widespread poverty, but this expectation is

convincingly disappointed by the historical accounts of

Captain James Cook and his crew during their sojourn in

the islands in 1778 and 1779, which describe Hawai‘i

prior to the massive changes eventually brought on by

contact with outsiders. This is the testimony of men –

there were no European women on board Resolution or

Discovery – who had spent several years in the Paciﬁc

and were intimately familiar with islands and islanders

throughout the Polynesian triangle. Lieutenant King

remarked that sweet potato, the pre-eminent rain-fed

agricultural crop in Hawai‘i, “thrives prodigiously, indeed

it is such Plenty that the poorest natives would throw

them into our Ships for Nothing” (Beaglehole 1967:

618). The sweet potatoes themselves were described by

James Trevenen, a midshipman on Resolution and later

Discovery, who judged them “inﬁnitely superior to any

others we ever met with at the Society, or Friendly

Islands . . . they are bigger than a Man’s head, sweet,

and mealy when dressed, & when raw taste something

like a chestnut” (Beaglehole 1967: 618, n. 1). The

testimony of King and Trevenen indicates that a

reasonable surplus was indeed available.

Of course, 1778 and 1779 might coincidentally have

been good years for the sweet potato crop, and a deviation

from the usual situation, but if this were so, it would be

difﬁcult to explain the large size of the pig herds raised

and managed by Hawaiians.

Captain Clerke referred to Kaua‘i Island as “the most

extraordinary Hog Island we ever met with, take them for

Number and size” (Beaglehole 1967: 575). Lieutenant

King’s corroborating assessment included a direct

comparison with Tahiti and the Society Islands:

Notwithstanding the much greater quantities of roots &

hogs that we destroyd, & of the latter salted down, than

at Otaheite or the Society Isles; yet here we never

perceivd this had any effect upon the great plenty still

on shore; . . .

Whereas at Otaheite the last time, when things were

found in the greatest plenty, Otoo was obliged to take

some pains in Supplying us. (Beaglehole 1967: 619)

This testimony to the large size of pig herds is bolstered

by native historians who refer to the sacriﬁce of large

numbers of pigs during rituals, as many as 800 or 2000 at

a time (Handy & Handy 1972: 252; Malo 1996: 252 ff.).

The link between the rain-fed agricultural ﬁelds and the

large pig herds described by Cook’s crewmen is clear –

sweet potato was the pre-eminent pig fodder throughout

Polynesia and Oceania:

It is in the singularly important role of conversion of

vegetable material into rich sources of protein and fat –

Archaeology in Oceania 3

animal husbandry – that the sweet potato attains

ascendancy over any other single plant species. Not

only is its ability in this direction recognized, but also,

in the feeding of pigs, there is a marked preference over

other cultigens. (Yen 1974: 52)

Sweet potato can be fed to pigs in several ways (Yen

1974: 52): leftover meal scraps; substandard roots from

harvesting; leaves and stems harvested especially for

fodder; after cooking; or by allowing pigs to forage ﬁelds,

either after ﬁnal harvest, or, in Hawai‘i, before harvest, a

prerogative accorded pigs as “bodies” of the god,

Kamapua‘a (Handy & Handy 1972: 253). At Kaupo¯, Maui,

rapid growing varieties of sweet potato were used

expressly as pig fodder (Handy 1940: 151).

In fact, the link between the rain-fed agricultural ﬁelds

and pig herds was so strong that an authority on sweet

potato agriculture in Oceania believed that the relationship

to pig husbandry deﬁned the place of sweet potato in the

traditional Hawaiian agricultural system: “[i]t is in the

analogies of implied surplus in both sweet potato and pig,

in its ceremonial use, gift exchanges, and ﬁnally, trade,

which allows for a thesis of agricultural development

whose course was accentual within the diversiﬁed use of

the environment” (Yen 1974: 312). The large Hawaiian pig

herds described in 1778–1779 were probably the result of

growth over a substantial period of time, a testament to

successful management of the variable fruits of rain-fed

agricultural ﬁelds.

THE ABSENCE OF ARCHAEOLOGICAL

EVIDENCE FOR BAD YEARS

Bad-year economics theorises Hawaiian society as living

on the margin, producing just enough to feed farmers and

ﬁnance ali‘i projects in good years and often falling below

this level so that ali‘i and maka‘a¯inana compete for scarce

resources (Figure 1). Population is a key variable in this

formulation, as it is for Malthusian theories in general,

because the situation where population outstrips resources

is theorised as an important engine for social change;

population size must be sufﬁcient to put pressure on

resources so that bad years are frequent. Both Kirch

(2010a) and Hommon (2013) take this bad-year economics

stance, and they both rely on population pressure to

promote social change, though they differ on the nature

and timing of that pressure. This is important because

population is difﬁcult to measure with archaeological

materials. In practice, what this means is that bad-year

economics accounts are based on rather weak

archaeological inferences about population history, which

they supplement with other evidence that is argued to be a

proxy for the relationship between population and

resources.

Archaeological inferences about population size are

weak because of the nature of the evidence. Although

Hawaiian archaeologists have repeatedly tried to estimate

population histories, they have yet to control adequately

for potentially confounding variables. The two approaches

that they have employed have been critically reviewed

(Clark 1988; Dye 2010b: 142–4; Kirch 2010a: 128–40).

The ﬁrst was the “house-count” method, which assigns

ages to habitation structures and counts how many houses

are present at different periods of time. The other,

sometimes referred to as the “Dye–Komori” method after

the authors of a paper that applied it to

C data from

Hawai‘i (Dye & Komori 1992), is based on an idea

developed by Rick (1987), which treats the corpus of

ages from habitation deposits as a sample from a charcoal

population that is argued to be isomorphic with population

size over time. In the years since these approaches were

developed, some fundamental problems with their

implementation in Hawai‘i have been identiﬁed. First was

the realisation that a failure to identify the charcoal

selected for dating meant that Hawaiian archaeologists

were often dating old wood. An early investigation in

Hawai‘i determined that old driftwood might add several

hundred years to the age of a dated sample (Emory &

Sinoto 1969); an attempt to measure its effect on

archaeological materials found an average error of about a

century in a sample of 40 age determinations (Dye 2000).

Subsequently, old wood was identiﬁed as a probable

cause of a 600-year error in age estimates for the early

O18 site (Dye & Pantaleo 2010). Old wood has a similar

negative effect on both the house-count and the

Dye–Komori approach to estimating population, in both

cases inﬂating estimates for early periods and deﬂating

estimates for later periods, yielding population curves that

overestimate the rate of early population growth and

magnify any slowdown of population growth later in the

sequence.

A second problem, identiﬁed somewhat later, has to do

with the failure of Hawaiian archaeologists to establish a

stratigraphic association between the dated event and the

target event. This creates a problem for the house-count

approach when a dated event is not associated with the

stratigraphic position of a house foundation. A review of a

systematic excavation program of 219 test pits through

surface architectural features at Kahikinui, Maui (Dixon

et al. 2000) found that most houses there had been built on

top of earlier cultural deposits containing charcoal. In the

median case, more than 80% of the cultural deposit

predated the excavated structure (Dye 2004). If, as is

commonly the case in Hawaiian archaeology, material for

C dates was collected without regard to the stratigraphic

position of the surface architecture, then it probably

derived from pre-construction deposits, thereby inﬂating

the age estimate for the house by an unknown number of

years (Dye 2004). The effect of the association problem on

population estimates derived with the house-count

method would be similar to the effect of old wood – an

overestimate of population in early periods and an

underestimate for later periods. The association problem

has a different negative effect on the Dye–Komori

method. Here, the problem has to do with identiﬁcation

of charcoal as deriving from a habitation deposit based

4 Wealth in old Hawai‘i

on characteristics of the surface architecture at a charcoal

collection locality; the lack of a demonstrated association

of the surface architecture with the dated material means

that there is no explicit reason to believe that the dated

material derived from a household context – the house

might just have well been constructed on an agricultural

or other non-habitation deposit that yielded charcoal for

dating. Because the dates used in the Dye–Komori

method must be conﬁdently associated with habitation

activities, it is not possible to argue convincingly that the

age distribution of charcoal not associated with surface

architecture is isomorphic to population size.

Despite these potentially grave and largely unresolved

problems with archaeological population estimates in

Hawai‘i, Kirch (2010a: 131–40) argues that when the

various population history estimates are taken into

account, a change in population growth around AD 1500

can be detected. In this view, a demographic transition

from high to low rates of growth at the beginning of the

sixteenth century was brought about because resources

could no longer support a growing population, a situation

with “enormous implications for sociopolitical change”

(Kirch 2010a: 139). Yet, the evidence for a change in

population growth at this time is problematic. The

argument relies on two early population history estimates

(Dye & Komori 1992; Hommon 1976), both of which

show a marked change to lower growth rates in the early

sixteenth century, but it ignores the fact that both of the

estimates are based primarily on unidentiﬁed charcoal

dates from contexts without a demonstrated association

with the surface architecture. Here, the errors probably

introduced by the failure to control for the problems of old

wood and stratigraphic association both depress late-period

population estimates, which would accentuate or create the

impression that population growth rates had changed.

Next, a plot of the number of conventional

C age

determinations by century for the islands of Hawai‘i,

Maui, Moloka‘i and Lana‘i, which in total show a slight

decline in the rate of growth beginning in the sixteenth

century (Kirch 2010a: 135), is offered as evidence for a

change to lower growth rates in the early sixteenth century.

However, this population estimation procedure is

problematic. No theoretical justiﬁcation is given for the

implicit claim that the temporal distribution of

C dates

from all contexts – habitation, agriculture, ritual and others

– is isomorphic to population size. There is also a

technical problem; the graphic purports to plot

conventional

C ages on the calendar timescale (Stuiver &

Polach 1977). If true, both the graphic and inferences from

it about events on the calendar timescale are nonsense –

C years are not calendar years and it is not possible to

get from one to the other without calibration. There can be

no assurance that the evidence cited for lower growth rates

in the sixteenth century is anything other than a reﬂection

of the uncontrolled effects of potentially confounding

errors.

Hommon recognises the limitations of archaeological

estimates of Hawaiian population (Hommon 2013:

205–12) and turns instead to a measure of population

growth that does not involve counting houses or tallying

residential charcoal samples. In their place, he argues that

the development of rain-fed agricultural features in the

Leeward Kohala Field System (LKFS) is, itself, evidence

for either a growing population or for “rising tax levies”

(Hommon 2013: 253). The features themselves do not

indicate how the increase in production they might have

effected was actually used, whether it kept the children of

farmers from going to bed with an empty stomach, or was

instead collected by ali‘i to ﬁnance projects. An increase

in production would move the state of the system to the

left on the production function graph (Figure 1), but the

starting position for the move is not known; it might be

near the bad-year section or the good-year section of the

function.

In this situation, archaeological evidence for common

bad years might indicate that development of rain-fed

agricultural features took place when population and

resources were out of balance. Elsewhere in the Paciﬁc,

archaeological evidence for common bad years includes

development of food storage facilities to buffer periodic

resource shortages.

One type of food storage facility is a pit, likened to an

“underground silo” (Kirch 1984: 133), lined with leaves

and used for the semi-anaerobic fermentation of breadfruit

and other starches, which is reportedly capable of storing

food reserves for decades. Pit-storage of starch staples is

known in both Western and Eastern Polynesia (Kirch

1984: 133), primarily on small islands where drought or

hurricanes periodically destroy crops and create famine

conditions. A few pits possibly used for this purpose have

been reported in Hawai‘i (Nakkim 1970), but decades of

archaeological investigation have not found other features

like them and it seems likely that they were either

extremely rare or absent.

Similarly, hilltop fortiﬁcations in New Zealand, called

pa¯, sometimes appear to have been designed to protect

food storage pits from raids by neighbours during bad

years (e.g. Fox 1976: 28–9; Kirch 1984: 209–11; Law

& Green 1972). Fortiﬁed food storage facilities such as

these have not been reported from Hawai‘i, whose

archaeological record includes only rudimentary forms of

fortiﬁcation. Thus, the archaeological record does not

indicate that bad years were common in Hawai‘i, nor

that rain-fed agricultural features were developed in

response to resource shortages. The alternative

hypothesis, that the rain-fed agricultural features were

developed in the context of status rivalry, is equally

plausible.

EQUIFINALITY AND INTERPRETATION OF

SYNCHRONIC DATA

One of archaeology’s strengths is its ability to yield

diachronic data capable of being interpreted as change in

human behaviour. The lack of archaeological evidence for

bad-year economics in Hawai‘i means that both Kirch and

Archaeology in Oceania 5

Hommon have to place great weight on a variety of

synchronic non-archaeological evidence, and that they

interpret this evidence as indicating an imbalance between

population and resources. A reliance on synchronic data

opens their arguments to the problem of equiﬁnality, where

different historical processes plausibly result in the same

outcome (Tuggle 2010).

Kirch and Hommon interpret elements of Hawaiian

culture and tradition as yielding evidence for purported

resource shortages in old Hawai‘i: Hawaiian culture

classiﬁes certain vegetable foods as “famine foods”, which

has led to the conclusion that famines were well known

and relatively frequent (Hommon 2010) and oral traditions

are interpreted to “refer more than once to periods of

drought, and to resulting famine, social disruption, and

warfare” (Kirch 2010a: 199; see also Hommon 2013: 238).

Both of these interpretations beg the question of how well

the Hawaiian word wı¯ translates as the English word

“famine”; that is, whether the range of historical events

classiﬁed as wa¯wı¯, or “times of famine”, included the

mortality and social disruption that accompanied famines

in Europe, such as the great famine in Ireland in

1845–1849, in which an estimated 1,000,000 people

starved to death when potato crops failed. A review of

Hawaiian oral traditions suggests that the meanings of the

two words overlap only partially.

The Hawaii Legends Index has 16 entries for “famine”.

These entries refer to four traditions, including the legends

of Kiha-a-Pi‘ilani, Hua, Kahuoi and Hina Kuluua. The

four traditions yielded 16 entries in the index because

some of them were retold widely in versions that refer to

different places and times. None of the famine traditions

mentions mortality, or relates resource scarcity to social

disruption or warfare. These characteristics appear to be a

product of their interpretation. In fact, a common theme of

the legends is decidedly non-Malthusian – access to food

was a right of person (Macpherson 1985) made possible

by a socially prescribed food-sharing ethic. This ethic was

expressed by “the accepted greeting, even to comparative

strangers”, which is translated literally as “come in and

eat, eat all you want” (Pukui et al. 1972), combined with

prohibitions against stinginess (Handy & Pukui 1972:

191). The legend of Kiha-a-Pi‘ilani (cited by Kirch 2010a:

199) tells how, after clearing a garden patch during a

famine, “Kiha-a-Pi‘ilani went to [the neighbouring lands

of] Hamakuapoko and Hali‘imaile to ask for potato slips.

The natives gave him whole patches of them wherever he

went. ‘Take a big load of the slips and the potatoes too if

you want them’ [they said]” (Kamakau 1992: 24).

The other legend cited as indicating famine is set in the

time of La‘amaikahiki, dated genealogically to the twelfth

century, a century or two after Polynesian settlement and

presumably well before the population could conceivably

have grown large enough to approach carrying capacity.

This legend, too, refers to the sharing ethic. It is set

during a spell of great drought, when a great famine

was experienced over all the lands from Hawaii to

Kauai, all the wetlands were parched and the crops

were dried up on account of the drought, so that

nothing even remained in the mountains. Waipio was

the only land where the water had not dried up, and it

was the only land where food was in abundance; and

the people from all parts of Hawaii and as far as Maui

came to this place for food. Because of this drought, all

the lands from Hawaii to Kauai were without food and

the people were forced to subsist on mosses [seaweed]

and other such things. But all through the drought and

famine Waipio never went without food. During this

famine the people from Hawaii, Maui and other islands

came to get food at Waipio. (Fornander 1916–1919: IV,

136)

The other two famine traditions also refer to this food-

sharing ethic. It seems perverse to interpret Hawaiian oral

traditions as indicating that resource shortages led to

social disruption and war, when in fact they celebrate a

right of person that would have counteracted these very

tendencies.

Similarly, the bad-year economics interpretation of a

famine tradition from Kona during the reign of

Kalani‘o¯pu‘u situates a cause of warfare in the tension

created by ali‘i demand for tribute and uncertainty in

producers’ ability to meet those demands (Hommon 2013:

238). This tension is indeed central to a peripatetic court,

which moves from place to place, eating from the fat of

the land (‘a¯ina momona). But just because this tension is

central to the relationship, social mores had developed to

ensure that it did not break out of the realm of

provisioning the king with food and supplies, and into

direct conﬂict between the people and the king’s court.

This would appear to be the message of the tradition about

Kalani‘o¯pu‘u. In Kamakau’s telling of the tradition,

Kalani‘o¯pu‘u is directly criticised as “senile with age”

(Kamakau 1992: 106) for his exorbitant demands, which

made life in Kona “uncomfortable” (Kamakau 1992: 105).

Here, Kamakau’s criticism of Kalani‘o¯pu‘u refers

implicitly to social mores concerning acceptable levels of

tribute demands with which his readers were expected to

be familiar. Although the tradition indicates that the Kona

people “wept bitterly” (Kamakau 1992: 105) at their fate,

the rebellion against Kalani‘o¯pu‘u that followed did not

arise in Kona, but in the distant district of Puna, where a

disaffected ali‘i usurped tribute that would customarily

have been claimed by Kalani‘o¯pu‘u. In any event,

Kalani‘o¯pu‘u effectively resolved the uncomfortable Kona

famine by moving his court to North Kohala, rather than

staying in Kona, where he risked being killed, as had other

ali‘i in the past who, for unspeciﬁed reasons, chose instead

to stay beyond their welcome. In North Kohala, a Ka‘u¯

chief named Nu‘uanupa‘ahu, who was travelling with the

court, joined the rebellion. He was found out by

Kalani‘o¯pu‘u and died soon after from a shark bite

suffered while surﬁng. This incident conveys supernatural

support for the reign of Kalani‘o¯pu‘u and reinforces the

message of the tradition that the locus of rebellion was the

6 Wealth in old Hawai‘i

relationship of the king to his nobles, who viewed signs of

his apparent senility as a weakness that might be

exploited. The tradition offers no support for the view that

warfare was sparked by strains in the relationship of a

king with his maka‘a¯inana. It clearly refers to the status

rivalry characteristic of Hawai‘i and other Polynesian

societies (Goldman 1970), in this case, rivalries that

sprang up soon after Cook’s death in 1779.

Two other pieces of synchronic evidence have been

interpreted to bolster the case for bad-year economics.

Evidence for depletion of soil nutrients in the sweet potato

plots of Hawai‘i and Maui Islands (Hartshorn et al. 2006;

Meyer et al. 2007; Vitousek et al. 2004) has been

interpreted by archaeologists as indicative of declining

yields associated with agricultural involution (Hommon

2013: 232–3; Kirch 2010a: 149–50). Soil scientists

interpret this evidence with a bit more circumspection; one

study concluded that rain-fed agricultural soils on Maui

produced crops sufﬁcient for local demands over very long

time frames (Hartshorn et al. 2006). On Hawai‘i Island, a

similar study has determined that agricultural practices

might have lowered yields at the upper edge of the LKFS,

where high rainfall had already leached nutrients from the

soil (Meyer et al. 2007), but soil nutrient levels within the

main part of the ﬁeld system, although measurably

lowered by traditional Hawaiian cultivation, were still

relatively high (Chadwick et al. 2013). Hawaiian farmers

made considerable use of mulch (Malo 1951: 205;

Menzies 1920: 76), which would have worked to reverse

deleterious effects of nutrient depletion. In addition to

ongoing investigations of mulching in traditional Hawaiian

agriculture (Lincoln 2013), soil nutrient studies might

model the effect of pig husbandry on sweet potato yields.

Pigs till, weed and fertilise, and the presence of a large

herd in the ﬁeld system might well have had a positive

effect on yields. At this point in the investigation of the

effects of traditional agriculture on soil fertility and crop

yields, it is premature to reject either a history that posits

declining yields or one that posits growing yields, both of

which should be considered as alternative hypotheses to be

tested by further research.

Similarly, women’s labour in the rain-fed agricultural

ﬁelds of Maui and Hawai‘i has been interpreted as a

response to labour shortage in the face of declining yields

(Hommon 2013: 61; Kirch 2010a: 139, 196, 216), but it

could just as well be interpreted as an integral part of the

intensiﬁcation process, unrelated to agricultural involution.

Polynesian women traditionally played a role in the

production of valuables and wealth-assets (Weiner 1992),

and this appears to have been the case in Hawai‘i, as well

(Hommon 2013: 237–8; Linnekin 1988, 1990). The labour

of women in the rain-fed agricultural ﬁelds was probably

tied, at least to some degree, to the care of pig herds

(Hommon 2013: 78; Kirch & Sahlins 1992: 28), and thus

might be interpreted as an expansion of women’s

traditional role as producers of valuables and

wealth-assets, rather than an indication that something had

gone wrong.

GOOD-YEAR ECONOMICS

Bad-year economics has developed increasingly complex

models that relate population and productivity under

various constraints (Lee & Tuljapurkar 2008, 2010; Lee

et al. 2006; Puleston & Tuljapurkar 2008; Tuljapurkar

et al. 2007). This potentially useful and interesting work is

at an early stage of development. The models developed so

far are necessarily simple ones that stint “cultural,

institutional, and political factors in population dynamics

. . . [even though] such factors are important, always and

everywhere, and presumably become even more important

as political complexity increases” (Wood 1998: 100). One

characteristic of these simple models is that the

relationship between population and resources has one

stable equilibrium state (Kirch 2010a: 194), which is in the

bad-year region of the production function where T and V

intersect (see Figure 1). Such models describe simple

societies, whose functioning must be different than a

politically complex society, like Hawai‘i, that is

remarkable for its elaboration of the Polynesian cultural

theme of status rivalry, invention of state institutions and

creation of an endogamous ali‘i political class. Theory

with these limitations cannot expect to cover the full

range of possibilities explored by traditional Hawaiian

society.

One particular limitation of bad-year economics theory

is that it divides production in two, distinguishing a basic

level of production to meet the minimum subsistence

requirements of the cultivators from surplus that can be

used to feed livestock, support non-producers or exchange

for one purpose or another (see Figure 1). Good-year

economics theory requires that the surplus category be

further divided into consumables, which contribute to

well-being and quality of life, and wealth-assets, which are

the foundation of power (Bell 2004: 12). In the case of the

rain-fed agricultural systems of old Hawai‘i, the division

of surplus into consumables and wealth-assets means

distinguishing between individual pigs and herds of pigs.

Individual pigs are, of course, consumables the fatty ﬂesh

of which was widely prized throughout Polynesia. But

herds of pigs are different.

When managed correctly, they satisfy the four criteria

that distinguish wealth-assets cross-culturally (Bell 2004):

(i) a capacity for growth; (ii) the ability to “generate a

ﬂow of consumption beneﬁts to a set of individuals who

have rights to them” (Bell 2004: 13, italics in original);

(iii) scarcity in the sense that increases in size or value

must be socially recognised and approved; and (iv) their

exploitability over an indeﬁnite time horizon.

Hawaiian pig herds were capable of growing in size,

theoretically limited only by the high natural growth rate

of the pig population. In practice, the growth rate of pig

herds would have been slower than this theoretical rate,

but robust rates of growth in the pig population appear to

have been achieved with some regularity, as the

testimonies of Captain Clerke and Lieutenant King about

the abundance of hogs, cited earlier, attest.

Archaeology in Oceania 7

Second, pig herds generated a ﬂow of consumption

beneﬁts, primarily to male elite, whose successful claims

of rights to control their distribution culminated in an

ideological project of unusual scope in Polynesia.

Considered a body of the god, Lono, pigs were sacriﬁced

in Lono ceremonies, but were also appropriate offerings to

other gods (Valeri 1985a: 45). They were the most

important sacriﬁcial animal (Valeri 1985a: 47), and as

such, were forbidden to women and were rarely consumed

by low-status men. Pigs were “eaten most often by

high-ranking men in temples to which male commoners

were not admitted any more than women” (Valeri 1985a:

127). These temple activities have been described as feasts

and “a distinctly chieﬂy form of consumption [where] . . .

foods, which in other Polynesian societies might be

reserved for special feast occasions, had become the daily

fare of ali‘i households” (Kirch 2001: 178).

Third, pigs appear to have been distributed as a highly

valued gift. When Cook stopped at Kaua‘i in 1778, before

the English had impressed on Hawaiians what they wanted

by way of provisions, the ﬁrst canoes to come from shore

to the ships brought piglets as gifts.

David Samwell, surgeon aboard Resolution, learned that

pigs “are what they present to Strangers as a token of

Friendship at the ﬁrst Meeting” (Beaglehole 1967: 1082),

a custom that Cook later speciﬁed was practiced by

“chiefs or people of Note” (Beaglehole 1967: 491).

Pigs were given to Cook and his crew many times

during their two stays in Hawai‘i, often during state

ceremonies that indicate the high value accorded to gifts

of pigs. One example is the ceremony performed by

Kalani‘o¯pu‘u, king of Hawai‘i Island, when he returned

from Maui to meet Cook at Kealakekua Bay. Three

canoes came off to the ships, and the third, with 18

paddlers, “was ﬁlled with hogs and various sorts of

vegetables” (Cook & King 1784: III, 17). A bit later, on

shore, after Kalani‘o¯pu‘u had given Cook ﬁve or six

feather cloaks, his people brought in “four very large

hogs, with sugar-canes, cocoa-nuts, and bread-fruit”

(Cook & King 1784: III, 17). Another example was an

important act in the reconciliation following Cook’s

death, when Kalani‘o¯pu‘u gave Captain Clerke 13 hogs

along with various vegetables following the return to the

English of Cook’s remains (Beaglehole 1967: 547). Gifts

of pigs appear to have been common outside these state

ceremonies as well. According to Clerke, “they had no

idea of making a present of a hog or two, but would

frequently beg your acceptance of ten or a dozen”

(Beaglehole 1967: 595) and he recounts a courtesy gift

of “thirty Hogs and as much Fruit as my people could

destroy in a week” (Beaglehole 1967: 595) after an

informal dinner with Kalani‘o¯pu‘u.

These observations indicate the high marginal utility of

pigs in old Hawai‘i – an increase in the size of the herd

would make possible more and larger gifts, which would

have had a positive valuation in Hawaiian society.

Fourth, the historical record makes it clear that pig

herds were exploitable over an indeﬁnite time horizon.

When Captain George Vancouver revisited Kealakekua

Bay in 1793, after Kalani‘o¯pu‘u had died, the new King

Kamehameha made a point of indicating to Vancouver,

whom he had met when Vancouver was a midshipman on

Discovery during Cook’s third voyage, that he had

inherited the pig herds controlled by Kalani‘o¯pu‘u. He

presented 90 “very large hogs” to Vancouver while

wearing “a printed linen gown, that Captain Cook had

given to Terreoboo [Kalani‘o¯pu‘u]; and the most elegant

feathered cloak that I had yet seen, composed principally

of beautiful bright yellow feathers, and reaching from his

shoulders to the ground on which it trailed” (Vancouver

1798: II, 126). In late eighteenth century England, a

“printed linen gown” referred to a loose, typically

full-length (Ribeiro 1984: 26) robe made of patterned

material in oriental fashion. It was often worn informally

by gentlemen instead of a coat, and was a favourite of

portrait artists at least since the time of Samuel Pepys,

who a century earlier had sat for a portrait in a gown that

signiﬁed that “he was a gentleman with a cultivated and

leisurely lifestyle” (Ashelford 1996: 102). Kamehameha

might have fancied the gown for much the same reason,

and he undoubtedly understood that clothing in England,

as in Hawai‘i, was a common sign of status (Tcherkézoff

2003:56). However, he would have worn the feather cloak

during this state ceremony as “a political statement . . .

that his relationship to others was of superiority of power”

(Kaeppler & de Rooij 2010: 39), in this case a superiority

that legitimised claims to the consumption beneﬁts of the

large pig herds of the island that had been held by

Kalani‘o¯pu‘u when Vancouver ﬁrst visited the islands.

Recent archaeological research in the LKFS has yielded

information on how and when pig herds became

wealth-assets in old Hawai‘i.

SWEET POTATO AND THE LEEWARD KOHALA

FIELD SYSTEM

Direct evidence places the introduction of sweet potato to

Hawai‘i some three to ﬁve centuries after Polynesian

settlement (Dye 2011a: 135). The plant underwent a

remarkable radiation after its late introduction. By the

early twentieth century, some 230 named varieties were

known (Handy 1940: 32–4), the products primarily of

rain-fed ﬁelds on the geologically younger islands

(Ladefoged et al. 2010) and colluvial sediments on the

geologically older islands (Kurashima & Kirch 2011;

Vitousek et al. 2010). The agronomic characteristics of the

plant opened up vast areas of the geologically younger

islands of Hawai‘i and Maui for cultivation (Figure 2) –

one estimate is that the agricultural potential of Hawai‘i

Island more than tripled and that of Maui Island almost

doubled with the introduction of sweet potato (Graves

et al. 2010: 157). A similar doubling of agricultural

potential was found on the geologically older island of

Moloka‘i, where the bulk of sweet potato was probably

cultivated on colluvial soils (Kurashima & Kirch 2011:

3664).

8 Wealth in old Hawai‘i

A sophisticated program of

C dating carried out at

agricultural features (Ladefoged & Graves 2008) and

temples (McCoy et al. 2011) within the LKFS (Figure 3)

makes it possible to track ﬁeld system development and

the tempo of temple construction using Bayesian

calibration methods (Buck et al. 1996). These methods,

which incorporate both stratigraphic information and

age determinations in the calibration, typically yield

results that are more precise than can be achieved with ad

hoc interpretative frameworks. In the case of the LKFS,

this precision exposes patterns of change that can be

interpreted as indicative of good-year economics.

Additionally, the gains in precision are sufﬁcient that the

archaeological data can be related meaningfully to the line

of ali‘i leading to Kalani‘o¯pu‘u and Kamehameha, starting

with ‘Umi a Lı¯loa, an ali‘i who ﬁgures prominently in the

traditions of Kamehameha (Valeri 1985b), and is believed

to have ruled in the early seventeenth century.

Dating of agricultural activities in the LKFS indicates

that the area was cultivated as early as the ﬁfteenth century

(Ladefoged & Graves 2008). Bayesian calibration of

dating samples collected beneath agricultural walls

indicates that facilities designed to increase production

were built in the less desirable southern part of the ﬁeld

system beginning in the late seventeenth or early

eighteenth century, and that they were developed primarily

in the eighteenth and early nineteenth centuries (Dye

2011b) (Figure 4).

These results of the Bayesian calibration differ from

the interpretation based on ad hoc dating methods in at

least two ways. First, they yield estimates of facility

construction dates the precision of which is speciﬁed by

the posterior distribution. In contrast, the ad hoc

interpretation of these data yields posterior distributions

for the various dated events (Ladefoged & Graves

2008: 782), but the age estimates of the target

construction events are vaguely speciﬁed – “[t]he

majority of the agricultural walls were probably

constructed after AD 1660, when four phases . . .

of building development occurred” (Ladefoged

& Graves 2008: 779).

Second, the Bayesian estimates indicate that

development of the ﬁeld system continued after European

contact. This ﬁnding has been disputed as an “incorrect

conclusion that the leeward Kohala ﬁeld system was abuzz

with activity in the early nineteenth century” (McCoy

et al. 2012: 1209). Yet, contrary to the bald assertion that

the Bayesian ﬁnding is “remarkable given the overall trend

of post-contact rural depopulation due to introduced

disease and migration to nineteenth-century port towns”

(McCoy et al. 2012: 1209), extant historical and

archaeological records indicate that the ﬁeld system was

inhabited well into the early historical period. A

missionary census in 1832, more than 50 years after

European contact, makes it clear that people were still

choosing to live in and cultivate the ﬁeld system.

The census counted 8014 people in North Kohala and

noted that “[m]any of these live along the western shore

where there is a good ﬁshing ground, a still greater

number along the line of cultivation which commences

two or three miles inland” (Adams & Athens 1994).

Similarly, a review of the history and cultural resources

of North Kohala characterised the period 1841–1863 as

one “in which settlements were consolidating on the

windward side and in the leeward uplands as the declining

Hawaiian population was withdrawing to more habitable

and attractive areas” (Tomonari-Tuggle 1988). The

transition to the historical period in leeward North Kohala

is marked archaeologically by stone-walled exclosures at

habitation sites, which Land Commission records suggest

were built in the 1840s to protect thatched houses and

kitchen gardens from feral cattle (Barrère 1983: 33).

Habitation exclosures marked by deposits of

mid-nineteenth century historical artefacts of

non-traditional materials such as metal, ceramic and glass

were recognised in the ﬁeld system in the late 1960s

(Rosendahl 1994), and have been identiﬁed subsequently

by numerous surveys throughout the region (Clark &

Rechtman 2004; Clark et al. 2010; Graves 1992; Graves &

Franklin 1998; Hammatt & Borthwick 1986; Loubser &

Rechtman 2007; Wulzen & Goodfellow 1995). This

appears to be a situation where a ruling theory that “the

dryland ﬁeld systems [in Hawai‘i] were all rapidly

abandoned within a few decades following European

contact” (Kirch 2010a) is almost certainly incorrect for the

LKFS and where model-based chronology building yields

a result consistent with the local historical record (pace

McCoy et al. 2012: 1209).

Figure 2. The distribution of irrigated wetland and

rain-fed dryland agriculture in the Hawaiian Islands. Note

the large areas of the Maui and Hawai‘i Islands suited to

rain-fed dryland agriculture.

Archaeology in Oceania 9

The posterior distributions of ﬁeld system facility

construction events yielded by the Bayesian calibration are

sufﬁciently precise to be plotted usefully within a

genealogical framework (Figure 4). Facility development

probably started in the ﬁeld system sometime after ‘Umi a

Lı¯loa. The long left tail of the posterior probability

distribution for Group 1 walls, which overlaps the time of

‘Umi a Lı¯loa, is due to their being the earliest facilities

established in the ﬁeld system; estimates of their

construction dates are not constrained by earlier dated

events. The peak of the posterior distribution is several

generations later than the genealogically reckoned time of

‘Umi a Lı¯loa. The Group 2 trails are the earliest extant

trails in the ﬁeld system. They run over the Group 1 ﬁeld

system walls and, in the detailed study area (Ladefoged &

Graves 2008), constrained the layout of two subsequent

phases of wall construction. Bayesian calibration places

Group 2 trail construction around the turn of the

eighteenth century, about midway between the time of

‘Umi a Lı¯loa and Kalani‘o¯pu‘u. The pace of construction

began to increase by the second half of the eighteenth

century and it is likely that the Group 3 walls and perhaps

the Group 4 trails were established during Kalani‘o¯pu‘u’s

lifetime. Facility development in the detailed study area

drew to a close when the Group 5a and 5b walls were

constructed an estimated 50–159 years (67% highest

posterior density region) after the Group 2 trails.

This combination of data and model indicates a rapid

development of facilities in the generations before

Kalani‘o¯pu‘u and continuing through to the early

nineteenth century and the time of Kamehameha.

A second line of chronological evidence comes from

estimates of temple construction dates in the uplands of

leeward Kohala near the agricultural ﬁelds. Although the

temple excavations are not fully reported, an initial

publication estimated the chronology of temple

development as having fairly great antiquity (McCoy et al.

2011), often in the face of relatively late

C dates from

contexts that predate temple construction (Dye 2012).

When the dating results are analysed using model-based

Figure 3. The location of the Leeward Kohala Field System. The shaded and labelled land units are where most of the

archaeological investigations of the ﬁeld system have been carried out.

10 Wealth in old Hawai‘i

chronology building, it is apparent that a series of mostly

small temples was constructed throughout the ﬁeld system

beginning perhaps as early as the sixteenth century, but

primarily in the seventeenth and eighteenth centuries,

when seven or eight of the 11 dated temples were probably

constructed (Figure 5). The boom in temple construction is

more closely associated with the time of Kalani‘o¯pu‘u than

it is the time of ‘Umi a Lı¯loa, although the dating

precision here is not as great as within the ﬁeld system,

where the wealth of stratigraphic information yielded by

the intersection of ﬁeld walls and trails improves the

calibration markedly. The contemporaneity of facility

development in the ﬁeld system and temple construction

identiﬁed by the Bayesian calibration is a pattern that was

obscured by the ad hoc chronology (McCoy et al. 2011),

which showed a gradual increase of construction activity

over a longer period of time.

The small size of many of the temples identiﬁes them

as family shrines, rather than community or polity-level

temples (Kolb 2006). Their construction appears to

indicate that maka‘a¯inana families had taken responsibility

for intensiﬁcation of production in the ﬁeld system and

that to some degree they subscribed to the ideological

project that institutionalised ali‘i claims to the

consumption beneﬁts of the pig herds raised there. If it is

true that pigs were typically eaten in temples late in

traditional Hawaiian times (Kirch 2001; Valeri 1985a:

127), then the temple construction sequence represents a

material result of social changes in the distribution and

consumption of pigs. The temple construction history

illuminated by the Bayesian calibration appears to indicate

that pigs were distributed for consumption ever more

widely during the eighteenth century, at least among

males, when the number of temple structures in which pig

was consumed grew dramatically.

If development of the LKFS does represent an effort to

create and maintain pig herds as wealth-assets, then it is

reasonable to ask if this development had any effect on

Hawaiian social organisation. The answer to this question

appears to be yes. Hawaiian tradition indicates that one

way in which ali‘i might invest pig herd wealth-assets to

expand their inﬂuence and elevate their status (Goldman

1970) was to ﬁnance alliances with powerful ali‘i families

on Maui Island and elsewhere (Abad 2000: 513–27,

570–2).

Accounts indicate that the frequency of marriage

alliances between Hawai‘i and Maui ali‘i begins to rise a

few generations before ‘Umi a Lı¯loa and continued

through the genealogical sequence, reaching a peak in the

early historical period during the time of Kamehameha

(Figure 6). Given the ceremonies of state reported by Cook

and his crew, in which large numbers of pigs were

exchanged as gifts, it is easy to imagine that marriage

arrangements among ali‘i families were made in the

context of gift-giving that included pigs. To the extent that

this was the case, the increased frequency of inter-island

marriage alliances attested by tradition might be seen as

both a cause and a consequence of the contemporaneous

development of large pig herds as wealth-assets in the

LKFS, as a source of power for the line of ali‘i leading to

Figure 4. The chronology of dated features in the Leeward Kohala Field System detailed study area. The vertical lines

indicate the approximate reign of ‘Umi a Lı¯loa in the early seventeenth century and the time of Kalanio¯pu‘u, who was

ali‘i nui of Hawai‘i Island when Cook visited.

Source: Dye (2011b).

Archaeology in Oceania 11

Kamehameha and one of the means they used to enhance

their status and expand their inﬂuence in traditional

society.

CONCLUSIONS

Bayesian chronologies of agricultural facility development

and temple construction in the LKFS, together with

traditions of ali‘i alliances kept by the Kamehameha line,

indicate that Hawaiian society was changing rapidly in the

generations before Cook arrived. Intensiﬁcation of

production in the ﬁeld system and the concomitant growth

of pig herds appears to have been accomplished by

families. The boom in construction of family-sized temples

in the seventeenth and eighteenth centuries indicates that a

family’s consumption beneﬁt claims against the herds of

pigs they managed were increasingly honoured in the

context of an evolving ideology that increasingly

Figure 5. The tempo of temple construction in leeward North Kohala during the post-settlement period. The vertical

lines indicate the approximate reign of ‘Umi a Lı¯loa in the early seventeenth century and the time of Kalanio¯pu‘u, who

was ali‘i nui of Hawai‘i Island when Cook visited.

Source: Dye (2012).

12 Wealth in old Hawai‘i

legitimated claims differentiated according to gender and

status. These patterns of change indicate a society with a

growing ability to produce wealth-assets. They contradict

bad-year economic theories (Hommon 2013; Kirch 2010a)

that posit a process of agricultural involution leading to

resource shortages and social unrest. Instead, they point to

a society in which maka‘a¯inana spent most of their lives

within the good-year region of the production function.

The idea that variable yields of rain-fed agricultural

systems in Hawai‘i were a cause of warfare that played a

central role in the development of social complexity has

been entertained by archaeologists for almost 30 years.

The theories developed from this idea have focused on the

possibility that bad yields not only reduced the amount of

surplus production, but also declined to levels that

threatened the subsistence needs of farmers. In this view,

tension caused by ali‘i demands for tribute from farmers

working at the edge of hunger led ali‘i to wage wars of

conquest against their afﬂuent irrigated agricultural

neighbours. Wide-ranging efforts to ﬁnd evidence for bad

years in comparative ethnography, Hawaiian tradition and

the results of multidisciplinary ﬁeldwork have yielded

little.

The reason for this difﬁcult search might be that

agricultural yields rarely, if ever, declined to levels that

threatened the subsistence needs of farmers. Bad years

were infrequent because Hawaiian society was operating

well within the good-year region of the production

function model (see Figure 1). However, good-year

economics does not contradict the idea that variability in

agricultural yield might have promoted warfare. Instead, it

suggests that ﬂuctuation of rain-fed agricultural yields led

to warfare because social relations among ali‘i families

were difﬁcult to maintain with unpredictable swings in

wealth (Goldman 1970: 204–12). Leeward ali‘i on Hawai‘i

and Maui, able to carry out wealth-intensive social

strategies during good years when pig herds were large,

found themselves unable to do so when crop yields and

herd sizes dropped. These ﬂuctuations, which must have

been a source of social friction in an alliance maintained,

in part, by the exchange of gifts, would at times have led

to wars, as one side or both came to see an alliance as less

desirable than expected and felt sufﬁciently aggrieved to

use force. This motivation for warfare might arise even in

times of relative plenty, and so would go some distance to

reconciling the Hawaiian traditions of warfare with the

Cook-era descriptions of a thriving and bounteous society.

In this view, Hawaiian ali‘i were not “hostile and

expansionistic” (Kurashima & Kirch 2011: 3673); they

were wealthy and inﬂuential.

The density-independent approach suggests two

archaeological projects for the LKFS. The ﬁrst builds upon

the work begun by McCoy et al. (2011) at the family-sized

temples. According to the preliminary description of this

work, excavations were conﬁned to small pits designed to

get under the temple foundation to recover material for

dating; these excavations provided data for the important

problem of when the temples were built, but did not

attempt to collect information with which to characterise

activities carried out at the temples. If it is true, as the

density-independent account holds, that the temples were

constructed as places where pigs were cooked and eaten,

then the question is raised whether the remains of those

meals can be found by archaeological investigation. A

research design for the temples would include a search for

earth ovens and trash pits outside of the temple structures,

perhaps accomplished by using a backhoe to strip off

pasture grass around one or more temples to expose a

broad area in which the tops of subsurface features might

be identiﬁed.

The theory predicts that trash pits associated with

temples will contain a high proportion of pig bone in

relation to other types of food remains. Comparative data

might come from earth ovens and trash pits associated

with nearby habitation sites, identiﬁed, recovered and

analysed in a comparable way. These excavations will

directly investigate maka‘a¯inana activities that contributed

to the development of pig herds as wealth-assets, and

might be interpreted as indicating the degree to which

maka‘a¯inana behaviour changed to support this ali‘i

project.

A second project to investigate the history of walled

habitation sites and the decline of the LKFS in the

nineteenth century incorporates both historical and

archaeological data. Land Commission records suggest

that the walls were put up around habitation sites by

people determined to stay on the land, but who were

eventually overwhelmed by the encroachment of feral

cattle and the growth of ranching, which transformed the

entire ﬁeld system to pasture in the last half of the

nineteenth century.

First-hand accounts of the difﬁculties faced by farmers

in North Kohala were probably published in

Figure 6. The generational frequency of inter-island

marriage alliances, Hawai‘i and Maui. The vertical lines

indicate the approximate reign of ‘Umi a Lı¯loa in the early

seventeenth century and the time of Kalanio¯pu‘u, who was

ali‘i nui of Hawai‘i Island when Cook visited.

Source: Graves, Cachola-Abad & Ladefoged (2010).

Archaeology in Oceania 13

Hawaiian-language newspapers of the day, and these might

be reviewed relatively easily because the work of

digitising Hawaiian-language newspapers of the era has

made substantial progress.

The historical record created

by this work can be supplemented by areal excavations at

walled habitation sites to chart changes over time in their

structure and the organisation of activities within them.

Theoretical and methodological issues in studying the

abandonment of settlements and regions have been worked

out by archaeologists (Cameron 1991; Cameron & Tomka

1993; Smith & Gazin-Schwartz 2008) and the insights

they offer might be applied productively to the situation in

North Kohala. Excavations at habitation sites can

document how daily routines were shaped by the

introduction of cattle and the extent to which the lives of

farmers changed before a traditional Hawaiian lifestyle

became impossible to sustain in the LKFS.

ACKNOWLEDGEMENTS

Figure 3 was drafted by Carl Sholin and Figure 2 by Eric

Komori, both based on data provided by Thegn

Ladefoged. Valuable comments on earlier versions of this

paper were kindly made by Jim Bayman, Geoff Boden,

Caitlin Buck, Tim Earle, Rob Hommon, Thegn Ladefoged

and Tim Rieth. An abbreviated version of the paper was

presented at a session honouring Pat Kirch at the 2013

Society for American Archaeology conference in

Honolulu.

The author is responsible for errors of fact and

interpretation.

COMMENT AND RESPONSE

PATRICK V. KIRCH

Departments of Anthropology & Integrative Biology,

University of California, Berkeley. kirch@berkeley.edu

Tom Dye provocatively advances several arguments: (1)

that a “bad-year” economic model fails to account for the

rise of Hawaiian pristine states; (2) that the Hawaiian

capacity to grow pig herds was “limited only by the high

natural growth rate of the pig population”; (3) that

Hawaiian population was not limited by agricultural

production capacity; (4) that pigs were the primary form

of consumable wealth utilised by ali’i families in their

power strategies; and (5) that commoner families “spent

most of their lives within the good-year region of the

production function”. I will comment – all too brieﬂy,

given the space allotted by the editor – on each of these

points.

1 Dye paints a caricature of the “density-dependent,

bad-year” economic model. In Figure 1, he positions

bad-year economics at the extreme end of the

production function (the equilibrium intersection of

variables V and T). I never claimed (nor do I believe

today) that the Hawaiian economy ever reached this dire

inﬂection point. When I ﬁrst presented the

production-function model, I theorised that in late

prehistory Hawai’i had entered a phase in which the

ability to produce surplus was declining in absolute

terms (Kirch 1984: 191, ﬁgures 50, 61). This is very

different from saying that people were starving every

year. Dye also erroneously pegs me as a strict

Malthusian, whereas in my recent book on the rise of

archaic states in Hawai’I, I am at pains to apply the

more nuanced “MaB rachet” model developed by Ron

Lee (1986), which integrates Malthus with Boserup

(Kirch 2010a: 190–201). Hawaiian farmers were not

helpless victims of a Malthusian crunch; they

continually innovated (and intensiﬁed), as so much

archaeological research has demonstrated.

2 Dye accuses me and others of depending on

“synchronic data”, yet the evidence he adduces for the

capacity of Hawaiian farmers to produce hogs relies on

limited accounts of European explorers. Captain Cook

certainly received unusually generous quantities of hogs

during his visit. But Dye fails to temper this evidence

with the unique circumstances of Cook’s encounter: (1)

that he was perceived, or at least promoted by the

priests, to be the arriving god Lono (Sahlins 1995),

patron deity of sweet potato and the dryland ﬁeld

systems; and (2) that his interactions occurred at

Kealakekua Bay, a royal centre inhabited by the

Hawai’i Island king and chiefs, who could draw upon

vast resources to assemble an unusual quantity of pigs.

Absent zooarchaeological data on pig production or

consumption in the hinterlands, there is no way to

know if Cook’s ad hoc observations have any relevance

to “normal” levels of pig production.

3 Dye claims that archaeological evidence for “bad years”

is lacking. If he means that there is no evidence for his

caricaturised view that people were starving every year,

I agree. But he dismisses too cavalierly, in my view, the

dramatic declines in soil fertility in both the Kohala and

Kahikinui agricultural systems documented by the

Hawai’i Biocomplexity Project (Hartshorn et al.

2006; Meyer et al. 2007). These are strong empirical

indications that nutrient draw-down negatively affected

yields. And, to suggest that pigs would “fertilise” and

have a “positive effect on yields” in the dryland ﬁeld

systems is nonsense, given that the pigs themselves

were being fed on the uptake from the same ﬁelds –

this was a closed system in that sense. Dye also fails to

cite the most recent demographic models of Field et al.

(2011), which point to the core of the Leeward Kohala

Field System (LKFS) having reached a

density-dependent state in late prehistory, while

population in the southern margins continued to grow.

4 Pigs were, without doubt, a signiﬁcant form of

consumable wealth in Hawai’i, primarily for the ali’i.

Indeed, Sahlins (1992) abundantly documented the

ceaseless demands for hogs made by the chiefs upon

commoners in Waialua, O’ahu in the early post-contact

14 Wealth in old Hawai‘i

era. But pigs were still a component of the “staple

economy” (since they had to be fed on agricultural

yields). Far more important to chieﬂy power strategies,

in my view, was the emerging “wealth economy” based

on birds’ feathers and featherwork (Kirch 2010a:

220–1), and quite possibly also on the control of other

durable goods, such as high-quality stone adzes. Dye

does not comment on this durable wealth of old

Hawai’i.

5 Dye would have us believe that late pre-contact

Hawaiian society possessed “a growing ability to

produce wealth-assets”. I agree that the highly

intensiﬁed dryland production systems of Kohala,

Kona, Kaupo¯ and other regions exhibited a remarkable

capacity to produce “wealth-assets” in the form of

staples, pigs, birds’ feathers and other commodities,

which were regularly collected and utilised by the

chieﬂy elites to further their political aspirations. I

disagree that this capacity was actually growing,

since much evidence suggests that the systems

were struggling to keep up with demand and that

absolute surplus may have been declining from

earlier peaks. Here, I concur with Dye that we

need more direct archaeological evidence (such as

quantiﬁed zooarchaeological assemblages from both

residential and temple sites) to test these alternative

propositions.

But we must also ask: For whom were these

“wealth-assets” being created? Who beneﬁted from the

long hours of back-breaking work that went into the

highly intensiﬁed ﬁeld systems of Hawai’i and Maui? For

the ali’i, no doubt, it was always a good year. As

Kamakau informs us, either the maka’a¯inana ponied up

the requisite quantity of tribute for their ahupua’a during

the annual Makahiki collection, or the chiefs’ warriors

would plunder the territory (Kamakau 1964: 20–1).

Without doubt, good-year economic theory is useful for

understanding chieﬂy power strategies. Whether it applied

equally to the “99%” of commoners is another question.

The “trickle down” of some pig meat at occasional temple

feasts does not mean that the maka’a¯inana were “living

high on the hog”.

––––––

ROBERT J. HOMMON

Archaeologist (hommon@comcast.net)

Comment Abstract

While in this paper Dye advocates good-year economics as

a superior explanatory approach to bad-year economics in

the study of the indigenous Hawaiian economy, he also

notes that the two approaches are not necessarily mutually

exclusive. My comment focuses on the latter point, ﬁrst by

suggesting that the Leeward Kohala Field System could

have supported large populations of both humans and pigs,

and second by showing that advocacy for either approach

may generate conﬁrmation bias.

Comment

Here, Dye presents another of his insightful contributions

to the theory and practice of archaeology in Hawai’i,

joining advances in chronometry, the nature of surface

architecture and the application of Bayesian statistics.

While his primary aim is to advocate good-year

economics as a superior explanatory approach to bad-year

economics, Dye points out that both approaches “would

seem to be required for completeness”, a view that I agree

will enhance our understanding of Hawai’i’s past. For

example, population growth and the development of large

pig herds need not and, on current evidence, cannot be

shown to be mutually exclusive explanations for the

development of the ﬁeld systems of leeward Kohala

District and elsewhere in Hawai’i.

By applying models employing data from studies of

various traditional Paciﬁc agricultural systems and other

sources (Hommon 2013: 58–63, 78–9), I suggest that the

Leeward Kohala Field System (LKFS) was probably

capable of supporting both a sizeable human population

and, at relatively little apparent cost to that population, an

abundance of pigs consistent with that evidenced in the

traditional histories and late eighteenth century visitors’

accounts.

For example, by estimates calculated by applying values

in these models, the yield of the sweet potato ﬁelds of the

6000-hectare LKFS, lying fallow 60% of the time, could

have sustained 22147 members of the district’s

maka’a¯inana and 500 members of its ali’i (chieﬂy) class, as

well as 5344 pigs, an average of about 1.5 pigs annually for

every six-member maka’a¯inana household. The pigs raised

in the system could have provided 167 pork-consuming

male members of chieﬂy families with 30 kg of pork per

month throughout the 8 months of the year when pork was

normally consumed. Management of the pigs would have

required the equivalent of 668 full-time pig-herders, or 3%

of the maka’a¯inana population.

The estimates generated here are, of course, presented

not as conclusions but, rather, to demonstrate a useful way

of exploring some of the variables relevant to the ﬁeld

system’s production. In my view, such thought experiments,

as long as assumptions such as fallow periods, caloric

content, crop yield and labour are culturally relevant and

speciﬁed, can help guide our thinking about how to study

the system’s use and development.

The statement by Earle (1978: 183), cited here by Dye

and elsewhere by various authors over the decades, that

agricultural intensiﬁcation “was an outcome of political

competition and not of population pressure” was based

on Earle’s research in the archaeology and ethnohistory

of Halele’a District, Kaua’i. He observed that “only the

prime areas of alluvial soils were farmed intensively

aboriginally. Even the alluvial soils were greatly

underutilized, as shown by the later, tremendous

expansion in irrigation agriculture for rice. The

Archaeology in Oceania 15

inattention to kula [rain-fed] farming and the restriction

of irrigation farming cannot be explained on

technological grounds” (Earle 1978: 163).*

Earle evidently assumed that had agricultural

development been generated by population pressure, the

residents of Halele’a would have extended cultivation

beyond the “prime” alluvial pondﬁelds by building dryland

ﬁeld systems similar to those of Hawai’i Island and Maui

to bring adjacent rain-fed areas under cultivation.

“Since this expansion was technologically possible, it

seems reasonable to conclude that it was not necessary.

For Halelea and most areas with irrigation, Hawaiian

agriculture does not show the intensive development

normally associated with population pressure”* (Earle

1978: 163–4).

Recent research by the Hawai’i Biocomplexity Group

has provided a more plausible explanation for the

absence of extensive traditional Hawaiian rain-fed ﬁeld

systems other than those on the geologically young

slopes of eastern Maui and Hawai’i and certain

rejuvenated environments. Slopes of the older islands,

including Kaua’i, having lost soil nutrients through

leaching, were too infertile for traditional Hawaiian

cultivation (Hartshorn et al. 2006; Vitousek et al.

2004). Earle (1978: 106–7) himself presents what

may be a practical explanation for the lack of

development of steeper alluvial lands in Halele’a

valleys. The Halelean farmers may have calculated that

terracing these slopes yielded insufﬁcient return on

investment.

“For a pondﬁeld of constant size, the volume of earth

moved in construction increases geometrically with the

natural slope of the land. By decreasing the average size of

a terrace on steeper slopes, however, the volume of earth

moved for a given area may be held constant. As the size

of a pondﬁeld decreases, however, the percentage of land

area devoted to bunds dramatically increases”* (Earle

1978: 106–7).

The farmers of Halele’a may have decided that the

potential beneﬁts of bringing relatively minor additions

into the ﬁeld system would not justify the costs of

construction and maintenance.

In somewhat the same way that this account of

Earle’s advocacy of the density-independent view

highlights the value of seeking alternative explanations

for observations (in this case the absence of evidence of

intensiﬁcation beyond the boundaries of pondﬁelds on

prime alluvial land), Dye’s discussion of the terms

“famine” and “famine foods” cautions us to exercise

care in translations and deﬁnitions of terms that may

seem to support the density-dependent view. While the

term “famine” usually refers to severe and long-term

lack of food resulting in high rates of death, it seems

clear that in Hawai’i before Western contact, while

certain normally unimportant food plants were exploited

during times of shortage of the major staple crops, such

events were usually limited in severity, duration and

geographical extent (Hommon 2010: 22).

*Corrections added on 11 April 2014, after ﬁrst

publication online: Quotation marks have been added to

text in the section above to attribute the quotes to the

referenced author.

––––––

ETHAN E. COCHRANE

Anthropology, University of Auckland, New Zealand.

e.cochrane@auckland.ac.nz

Dye’s analysis of social and political complexity in Old

Hawai‘i exempliﬁes the turn to an increasingly

well-founded evolutionary archaeological science of

Paciﬁc Island populations (e.g. Allen 1996; Field 2004;

Ladefoged, Lee & Graves, 2008; Lipo et al. 2010).

Although his paper is not a programmatic call, the

manner in which Dye has carefully analysed

data-generation procedures, his concern with the links

between models and observations, and his fundamental

reliance on concepts of competition for resources (status,

pigs, arable land) indicates that he is concerned with

both science and evolutionary change. Here, I want to

brieﬂy discuss two fundamental characteristics of

evolutionary science (indeed any science), both of which

play a role in Dye’s analysis, and argue for their

continued development and use in the archaeological

study of sociopolitical complexity in the Paciﬁc Islands,

and Paciﬁc archaeology generally. Undoubtedly, others

will be more concerned with substantive observations that

challenge Dye’s account of sociopolitical change. These

debates about data are important and necessary, but they

are insufﬁcient without the theoretically justiﬁed and

empirically relevant framework that evolutionary science

provides.

Two uncontroversial, and therefore fundamental,

characteristics of science are (1) the use of universal

concepts and mechanisms (theory) to explain empirical

phenomena, and (2) that observations of phenomena are

made with units linked to theory, and unambiguously

applicable to the empirical world. The best scientiﬁc

explanations match empirical observations to expectations

derived from theory or cannot be falsiﬁed. Although this

is an exceedingly simple characterisation, many will note

that it is different from the typical social science

approach, where observations may have no necessary link

to theory (they often come from our common sense) and

empirical observations are explained by subsuming them

under an existing empirical generalisation. Willer and

Willer (1973) call this approach to research “systematic

empiricism”.

Empirical generalisations, however, are often a

necessary starting point, as they tell us about possibly

important patterns in the world. Dye begins his paper by

describing the empirical generalisations used by Kirch

(2010b) and Hommon (2013) in their recent analyses of

Hawaiian sociopolitical complexity. These authors each

assemble a series of archaeological, ethnohistorical,

16 Wealth in old Hawai‘i

palaeoenvironmental and other observations to correlate

changing Hawaiian population levels with the amount of

subsistence resources. For Hommon and Kirch, the

occasional imbalance between relatively high population

levels and inadequate subsistence resources is a key driver

of sociopolitical change in what Dye terms a

density-dependent or bad-year economics approach.

Dye’s critical review of the observations that are the

foundation of the population-resource generalisation

underscores the importance of how we construct

observational units in an archaeological science. Dye notes

that “inferences” of ancient population size are problematic,

as there is no unit by which we can unambiguously observe

population size in the archaeological record. Neither

archaeological house counts nor radiocarbon date

frequencies (the Dye–Komori method), as typically

generated, allow us to observe population size in a manner

sufﬁcient for evaluating potential explanations in a

scientiﬁc framework, regardless of the use of these units to

make general, relative comparisons. As an alternative to

Kirch and Hommon, Dye constructs an empirical

generalisation that correlates changes in dryland farming

intensity, ali‘i genealogies and temple construction,

primarily family shrines. As he suggests, the observational

units that Dye employs are arguably more directly linked to

the archaeological record than units used to calculate

population size. They are therefore better from a scientiﬁc

standpoint, but there is still room for reﬁnement. For

example, what are the necessary and sufﬁcient

archaeologically observable criteria for identifying family

shrines within the ﬁeld of surface stone remains?

Theoretically informed artefact classiﬁcations are necessary

to address questions such as this.

In addition to building appropriate classiﬁcations or

observational units, universal concepts and mechanisms

are used to explain phenomena in a scientiﬁc archaeology.

In contrast, the explanations typically offered for changes

in ancient Hawaiian sociopolitical complexity are

empirical generalisations, founded on principles of

competition for resources, but lacking a clear mechanism.

For example, Hommon (2013) proposes that sociopolitical

change in ancient Hawai‘i is explained by chiefs asserting

greater control over subsistence practices during times of

hardship. This explanation is a product, in part, of

subsuming the observations of the Hawaiian

archaeological and ethnohistorical record under another

empirical generalisation correlating twentieth century

Tikopian chieﬂy practices and historically documented

environmental stress. We may ask, though, why are these

apparently status-linked behaviours differentially

transmitted and why do they persist in multiple times and

places? What is the mechanism? Here, human intention

and agency is often thought to offer a mechanism, but

these are problematic as explanations in evolutionary

science (Mesoudi 2008; Ormerod 2005). Although Dye’s

good-year economics model articulates better with

observations of the archaeological and ethnohistorical

records, it relies in part on determining the intentions of

individuals in the past to explain the course of history; for

example, that “ali‘i might invest in pig herd wealth-assets

to expand their inﬂuence and elevate their status” (second

italics not in original).

The explanations proposed by all of these authors are

founded on concepts of competition for resources,

concepts that are already part of evolutionary theory, and it

is from here that we should derive our explanatory

mechanisms. For example, Ladefoged et al. (2008) discuss

the probable intentions of chiefs in the development of the

Leeward Kohala Field System, but the universal

explanatory mechanism that shapes the distribution of

chieﬂy behaviours is selection, irrespective of an

individual chief’s intentions.

––––––

TIMOTHY EARLE

Anthropology, Northwestern University.

tke299@northwestern.edu

Tom Dye’s paper provides provocative insight into

the importance of pigs in the political economy of

protohistoric Hawai’i. The historical sources summarised

by Dye stress the abundance of pigs probably as the result

of extensive dryland sweet-potato farming there. His paper

has been chosen for commentary because the signiﬁcance

of this observation was seen as controversial. My

commentary builds on the history of Oceanic archaeology

dealing with agriculture to suggest why the paper might

make some scholars mad, but how in fact it suggests a

reconciliation between two materialist traditions that have

often been seen in conﬂict.

Dye’s paper refocuses our attention on one of the great

themes in Oceanic anthropology and archaeology – the

relationships between agricultural systems and the

development of complex societies. He describes two

approaches – density-dependent (what he calls bad-year

economics) versus density-independent (good-year

economics) models. These two approaches can perhaps

best be understood in terms of the history of materialist

studies in anthropology. The “density-dependent”

approaches derive from ecological anthropology, as

articulated by Julian Steward in the 1950s and represented

in archaeology through settlement pattern studies

introduced by Roger Green to Paciﬁc archaeology in the

1960s. Settlement pattern studies stress the long-term

adjustment of societal organisation to changing subsistence

challenges involving growth in population and agricultural

intensiﬁcation. The “density-independent” approach

derives from political economy approaches articulated by

V. Gordon Childe and ultimately from Karl Marx. These

emphasise surplus mobilisation and use in power strategies

in the creation of social inequality.

Historically, these two approaches have been set against

each other, largely because of Marx’s hatred for Malthus’

blame-the-victim argument that poverty was caused by

Archaeology in Oceania 17

over-breeding by the lower classes. More recently,

Friedman (1974) labelled those working with ecological

models as “vulgar” Marxists, creating a modern schism in

materialist approaches, with resentment among those using

an ecological approach. Dye’s paper suggests a means of

rejoining the two approaches as they refer to the role of

agricultural production in protohistoric Hawai’i.

In Social Stratiﬁcation in Polynesia, Marshall Sahlins

(1958) originally framed eclectically the relationship

between agricultural productivity and social complexity.

Sahlins sought to explain the extent of social complexity

as a result of adaptive radiation to diverse island

environments (from atolls to large volcanic islands), and

he combined the theoretical approaches of his two

professors Steward (at Columbia) and Leslie White (a

maverick Marxist at Michigan). Because of his creativity,

Sahlins forged a corrective synthesis between the opposed

positions of ecological and political approaches

represented by his professors. Subsequently, however, he

shifted ﬁrst towards a more ecological approach articulated

by his collaborator Elman Service and then towards a

more structural Marxist approach. His extremely articulate

(but shiftless) work has caused a widening in the schism

of materialist approaches based on internal contradictions

in his arguments.

Subsequent to Sahlins’ seminal book, and with the

introduction of settlement approaches à la Steward, the

ecological approach has become foremost in Paciﬁc

archaeology. The ﬁnest exponent has been Patrick Kirch.

In terms of agricultural studies in the Paciﬁc, Barrau

distinguished between wet (irrigated) and dry farming.

Contrasting hydrologic conditions available on islands

channelled these alternative farming strategies with

probable outcomes for social evolution. In The Wet and the

Dry, Kirch (1994) then argued that the relative unstable

conditions of dryland farming created intense warfare as

population growth caused increased competition over

limited and risky resources. In collaboration with Kirch,

Ladefoged et al. (2010) have conducted the most

inﬂuential study of the varying pathways (irrigated versus

dryland farming) that supported the emergence of

Hawaiian chiefdoms and subsequent states. They

demonstrate that across the Hawaiian Islands, a remarkable

contrast existed between the western islands where wet

(irrigated taro) farming dominated and the eastern islands

where dry (mostly sweet potato) farming dominated. Their

modelling showed that the dryland farming was both less

productive and higher risk than irrigation farming. To

Kirch (2010a), the increased risk of dryland farming on

the eastern islands helped explain why pristine states

emerged there and not on the more productive zones of

irrigation that characterised the western islands. This

ecological approach emphasises how polities emerged to

solve challenges (bad-year economics).

The “alternative” political economy approach is

represented in the Paciﬁc most clearly by Matthew Spriggs

and myself. Sahlins recruited me to show that Wittfogel’s

functionalist argument was wrong for the Hawaiian case

(Earle 1978, 1980, 2012). This was easy: the Hawaiian

irrigation systems on Kaua’i were not large scale (a

necessary condition for Wittfogel) and so did not require

central management. Rather, the engineered landscapes of

irrigation were highly productive (capable of producing a

surplus), sustainable and easily controlled by an emergent

elite. For Vanuatu, Spriggs (1985, 1986) emphasised that

erosion caused by dryland farming created the build-up of

alluvium that allowed for intensiﬁcation with irrigation.

Thus underlying ecological conditions created the potential

for controlled surplus production that fuelled political

elaboration (good-year economics).

Are the two materialist approaches in opposition? I do

not think so, and Dye makes the implicit argument that

they are complementary. To a large measure, he agrees

with the signiﬁcant scholarship on the dryland ﬁeld

systems conducted by Kirch’s collaborative group, but the

value added by Dye’s analysis brings in an explicit

political economy perspective. The introduction of sweet

potato to the Hawaiian Islands provided an added means

of intensifying agricultural production to support both an

expanding population and an expanding political economy.

Although less productive and higher risk than irrigated

taro farming, sweet potatoes provided the ideal food for

pigs that were a moveable wealth for political marriage

and sacred economies. Although lower in productivity and

higher in risk, the sweet potato ﬁelds thus supported the

production of a key moveable wealth; unlike taro, pigs

could be moved regionally to support political strategies at

some distance.

No topic could be as timely, given the recent

high-proﬁle publications about Hawaiian prehistory

(Bayman & Dye 2013; Hommon 2013; Kirch 2010a).

Dye’s paper demonstrates the importance of pigs as wealth

in the political economy, which has been largely missed in

previous studies. He provides a clear example of how

ecological approaches can be linked to political economy

approaches to give a powerful understanding of social

evolution. Why did the irrigation-based chiefdoms on the

western islands not develop into states? Probably, it related

to limits on scale for the smaller islands. The development

of moveable wealth based on the raising of sweet potatoes

appears to have allowed the spatially much larger polities

based on military conquest and ideological elaboration

(Kirch 2010a). In simple terms, the two approaches of

bad- and good-year economies are complementary. The

old nineteenth century controversies can be overcome to

create integrative, materialist theories for social evolution.

––––––

ROSS CORDY

University of Hawai’i – West O’ahu, Kapolei, Hawai’i.

rcordy@hawaii.edu

The pre-European formation of the Hawaiian kingdoms or

archaic states has been of archaeological interest since the

18 Wealth in old Hawai‘i

1970s, when Tim Earle, Rob Hommon and myself were

doing our theses on this topic (Cordy 1974a,b, 1978,

1981; Earle 1973, 1978; Hommon 1972, 1976). There

has been work since then by Hommon, Earle, myself, Pat

Kirch, Jane Allen, Michael Kolb, Boyd Dixon, Dennis

Gosser, Kehau Cachola-Abad and others. Kirch and his

colleagues have recently studied aspects of this topic,

focusing on rain-fed agricultural ﬁelds in leeward Kohala.

There are the recent books by Kirch (2010a, 2012) and

Hommon (2013), and also some key papers by my

colleague Tom Dye (2010a, 2011b), to which we can

now add this paper.

This is a topic that focuses on changes in political

organisation within polities over time, with the rise of elite

control. Over the years, different hypotheses have been

generated on what changed in political organisation and

when, and on what associated cultural variables may have

caused these changes. There is still disagreement on when

political organisation changes occurred (see Cordy 2012,

2013), and I would argue that no causal hypotheses have

been successful evaluated.

Tom’s paper brings up some important points. He

disagrees with the causal hypothesis of Kirch and

Hommon that population increased to levels approaching

production limits in rainfall-ﬁeld systems of Maui and

Hawai‘i, threatening available surpluses for those chiefs,

leading the chiefs to undertake conquest warfare, resulting

in social complexity. Tom points out that evidence cannot

yet be marshalled to support this hypothesis. He

emphasises that there is no sound archaeological evidence

(a) for population levels approaching production limits, as

population histories have not yet been accurately

reconstructed for any point in time prior to contact, and (b)

for indications of bad ﬁeld years. He suggests that oral

historical evidence for famine is rare (listing cases) and

does not show resulting mortality, social disruption or war.

He clearly doubts that bad agricultural years were

frequent, or threatened subsistence needs. Also, his

Bayesian dating for Kohala ﬁeld intensiﬁcation suggests

that it occurred after key political changes (also Dye

2011b: 28–31). These are telling points that others must

consider.

Tom proposes status rivalry as the causal factor

(independent of population growth), with wealth being

accumulated by chiefs for use in status rivalry, this wealth

including chieﬂy pig herds. He suggests that ﬂuctuations

in this wealth prevented chiefs from maintaining alliances,

and led to conquest and social complexity. He gives a few

examples for contact-era consumption of large numbers of

pigs at chieﬂy feasts and of unusual amounts of pigs

supplied to Cook’s expedition. While I agree that pigs

were important wealth items used by chiefs in the era of

concern, I believe that there is little understanding of pig

use at contact and even less evidence from pre-contact

archaeological reconstructions. Pigs were probably

differentially consumed in the diet of commoners, low

chiefs, high chiefs and rulers; used in differing numbers in

special feasts by these strata; and offered up in differing

numbers at the religious places of these strata. They were

also used to provision armies. None of this information

has been carefully compiled from the historical sources of

the contact era, nor how pigs were used in status rivalry

(no evidence has yet been shown that they were used in

marriage alliances), nor the nature of chieﬂy pig herds

(herds are rarely noted, and it is unclear how many of the

pigs supplied to foreign ships were from chieﬂy pig herds

or from demands on commoner households). Even if we

improve this contact-era picture, archaeology will have to

determine amounts of pigs at different points in time that

were used by chiefs for feasts and rituals; and this will not

be an easy task, as excavations in houses and religious

sites are usually limited, and recovered pig bone is often a

few fragments (e.g. Kolb, Conte & Cordy 1997; but see

Kolb 1999). Also, Tom’s claim that proxy archaeological

evidence of increasing dryland ﬁelds and small family

temples directly reﬂect increasing pig herds used for status

rivalry seems shaky. Tom cites Kaua‘i as a major pig

island based on Cook expedition journals, and my work

long ago showed that O‘ahu became the major

pig-producing island of the early 1800s (Cordy 1972).

Both islands had irrigated taro farming as their dominant

cultivation approach, suggesting that pig numbers’

relationship to the proposed proxy variable of

rainfall-ﬁelds is not so straightforward. Also, commoner

temples do not seem to have a tie to pig wealth of chiefs;

rather, actual pig offerings at major temples of high chiefs

and rulers seem more relevant. Kolb’s study of Maui

national temples, which suggested two periods of major

pig bone offerings (1500–1650 and 1650–1820), is a more

relevant ﬁnding (Kolb 1991, 1994, 1999).

Tom’s points about the inability to measure variables in

Kirch’s and Hommon’s hypotheses are very important

ones, and they seem to apply to Tom’s hypothesis too –

and to nearly all our hypotheses offered to date. We can

generate interesting hypotheses about what changed and

why, but we need to diachronically measure key cultural

variables of the hypotheses to evaluate them. If one is

using archaeological data, such measurement is extremely

difﬁcult. It is true for reconstructing population at different

points in time, as Tom notes (other issues on measuring

population exist; e.g. Cordy 2007; Kirch 2007). It is also

difﬁcult to archaeologically reconstruct levels of

agricultural production and surplus (as Tom notes) and

levels of pig production and surplus. It is true for

reconstructing changes in political organisation (number of

social strata, number of administrative levels of chiefs, the

end of kin-controlled local chiefs and land, etc.), and for

associated religious patterns (changes in the size and

offerings at national temples). Many of us have made

contributions to archaeologically reconstructing these

cultural variables – myself and Tainter on identifying

political hierarchies or strata (Cordy 1981, 2001; Tainter &

Cordy 1977), Dixon and Gosser on the possible end of kin

control in local leadership (Dixon et al. 2008; see also

Cordy 2004), Tom himself on the end of kin control of

resources (Dye 2010a), Kolb (1994, 1999) on changes in

Archaeology in Oceania 19

national heiau size and associated offerings, Allen on

irrigated agriculture (Allen 2000), and Kirch’s folks on

dryland ﬁelds and soil depletion (cited by Tom). But we

are a long way from archaeologically measuring these

variables at different points in time – identifying key

political organisation changes, measuring associated

cultural variables (population size, pig production, temple

sizes, crop production levels, warfare etc.) and determining

if key changes in these variables precede the political

changes, and are perhaps causal, or if they follow the

changes.

One of our biggest problems is ﬁne chronological

dating, as Tom and others have noted. I believe that Tom’s

use of Bayesian dating will prove useful in reﬁning our

chronologies. So too will the uranium–thorium coral

dating of some religious structures that Kirch has

introduced. Resolving the problems of hydration dating

could also help. Until we have more accurate dating, we

have problems studying this research question.

Yet another problem exists. In this research question,

the community (ahupua‘a) or region (moku such as

Kohala, Wai‘anae) are not the focal area. The changes are

in political organisation of a polity. Our reconstructions of

variables must not lose sight of what was the polity

pattern. We have only recently started to look at regional

or moku patterns (e.g. in Wai‘anae, Kula, Kahikinui and

Kohala), and the diachronic picture of key variables is far

from complete for any of these moku. Yet patterns of one

moku do not necessarily equal the patterns of its polity.

Our regional studies so far are in lands that generally only

had resident commoners and local chiefs. Research has not

focused on lands that were political centres (where rulers

and high chiefs lived). We have bits of information for

these areas from contract archaeology and some research

work at best.

Indigenous oral history is another major data source

that can address this problem. I disagree with Tom; oral

history is diachronic. It provides details for pre-European

times on polity sizes over time, variables of political

organisation, warfare patterns and so on. However, it too

has complexities as a data source that have to be carefully

dealt with. To note a few, some oral histories were

probably modiﬁed in pre-European times. Others recorded

in the 1800s may tell their stories with elements of contact

culture that did not exist in the past. Also, oral histories

recorded from contact up into the 1860s are more accurate

accounts at European contact; more recent oral histories

blend the earlier accounts and alter them, and must be

used much more carefully. Also, oral history goes back

accurately only to about AD 1300, and in the 1300s it is

spotty. Additionally, one has to be very careful not to

over-interpret some of the material. I have been critical

about weaknesses in the oral history presentations of Kirch

(Cordy 2012, 2013). Nonetheless, I have long found this

data source an extremely powerful tool to study this

research problem and measure key variables.

In sum, I ﬁnd Tom’s points on pigs as wealth important

to consider in the development of Hawaiian political

organisation, as I ﬁnd Kirch’s group’s Kohala work on

dryland ﬁelds. However, until we can diachronically

measure the variables of these hypotheses, they are only

fascinating ideas.

Last, I agree that status rivalry is involved in these

changes, or the desire for more mana. However, status

rivalry existed in the smallest of Polynesian polities

(e.g. Marquesas, Maori). For Hawai‘i, we need to

determine what associated factors led to the formation

of bigger polities and elite control. This clearly needs

much more work, and creative ideas such as this paper

of Tom’s.

––––––

THEGN LADEFOGED

Anthropology, University of Auckland.

t.ladefoged@auckland.ac.nz

Dye presents a theoretical dichotomy between bad-year

and good-year economic approaches.

He characterises density-dependent bad-year economics

as stressing variable yields in rain-fed agricultural systems

leading to food shortages and reduced tribute, with a result

being the stimulus of conquest warfare with more afﬂuent

irrigated agricultural areas. This characterisation is a

caricature of the approach, and misses several important

elements, including the insight of Kirch (1994) that it is not

just rain-fed variable production that is the stimulus but,

rather, the differential in reaching the inﬂection point of the

intensiﬁcation curve in rain-fed versus wetland areas. As an

alternative, Dye suggests that Hawaiian society was

generally based on a good-year economy, with agricultural

yields rarely, if ever, dropping below the subsistence needs

of farmers. In this view, everyone had enough food but

variation in production would have impacted on the elite’s

ability to engage in wealth-intensive social strategies such

as gifting to reinforce marriage alliances. It was apparently

this tension and the frustration over the ability to give and

receive gifts that led to warfare. Dye maintains that there is

little archaeological evidence to support density-dependent

approaches, whereas there is ample evidence for a well-fed,

wealthy elite-led, Hawaiian society engaged in periodic

warfare. As speciﬁed, Dye’s good-year economic

explanation is unsubstantiated and far too simplistic. In an

effort to distinguish it from bad-year economic approaches,

Dye creates a dichotomy between the two that is

overdrawn. Elements of structuralism and praxis are

currently incorporated into the bad-year economic

approaches of historical ecology and ecodynamics. The

empirical foundations of these approaches, including the

archaeological (see Kirch et al. 2012 for a summary),

geochemical (see Vitousek and Chadwick 2013 for a

summary), ethnohistorical (Abad 2000; Graves et al. 2010)

and experimental (Kagawa & Vitousek 2012) data, are

currently much ﬁrmer than those of Dye’s suggested

alternative.

20 Wealth in old Hawai‘i

Dye’s conclusions about the timing of agricultural and

religious development within the Leeward Kohala Field

System (LKFS) are based on his Bayesian analyses (Dye

2011b, 2012) of previously published data (Ladefoged &

Graves 2008, 2010; McCoy et al. 2011). Dye claims that

Bayesian calibrations have the potential to revolutionise

our archaeological interpretations, yet because his

Bayesian results are so sensitive to the assumptions of his

model, there are limitations. For example, Dye (2011a: 28)

makes the ad hoc assumption that the temporal boundary

for the last period of agricultural construction in the LKFS

should be based on a “normal curve with a ten year

standard deviation centred at AD 1850”. As noted by

McCoy et al. (2012: 1208–9), “the effect of choosing AD

1850 . . . [is that] . . . it forces the posterior probability to

ramp up, making it appear as if all activity in the ﬁeld

system occurred in the years immediately prior [to AD

1850]”. Unfortunately, the fundamental assumption in

Dye’s Bayesian model that agricultural development

continued until AD 1850 skews the results of his model

towards the historical period. Furthermore, the results of

Dye’s (2011b) Bayesian analysis are heavily inﬂuenced by

a strict adherence to the original ﬁeld recording and

analysis of Ladefoged and Graves (2008). If the original

recording of the spatial relationship of a single wall (in

this case, wall “c” intersecting trail “A”, as labelled by

Dye 2011b: Figure 2) was erroneous, it would signiﬁcantly

undermine Dye’s conclusion that much of the

intensiﬁcation of the LKFS is a post-contact phenomenon.

Given the inherent ambiguity of some ﬁeld decisions about

whether a wall intersects or abuts a trail, it is inappropriate

to create a dating model and derive conclusions about the

intensiﬁcation of the entire LKFS based on a method that

can be signiﬁcantly inﬂuenced by the misidentiﬁcation of a

single wall. While Bayesian analysis can contribute to our

understanding of past behaviours (e.g. Field et al. 2011),

the scale of the data and the assumptions of any model

must be closely monitored to achieve anything close to

revolutionary insights.

Dye places a considerable emphasis on pig husbandry,

yet he is not the ﬁrst to identify the importance of this

activity and the generation of surplus wealth in the LKFS.

Lockwood’s (2009) detailed analysis of lipid biomarkers

and XRF of elemental nutrients in soils collected from

various architectural structures in Lapakahi identiﬁed

spatial and temporal trends in pig husbandry. Lockwood

(2009: 242) concludes “that growth in the scale of surplus

production (as measured by pig husbandry) lagged behind

growth in the scale of agriculture and subsistence

requirements”. While acknowledging the limitations of his

analysis, Lockwood suggests that greater growth in pig

husbandry occurred during the earlier periods of

occupation as opposed to during the late prehistoric and

early historical eras. Dye’s citation of synchronic

traditional and historical accounts noting the importance of

pig production for wealth-assets is useful and limited, and

Lockwood’s analysis is an excellent ﬁrst step in

establishing diachronic trends within the LKFS.

An emphasis on good-year economics is an interesting

alternative for understanding Hawaiian sociopolitical

development in general, and developments within the

LKFS speciﬁcally. However, it would seem that in his

enthusiasm to distinguish the perspective, Dye has

overstepped the data. On the basis of his Bayesian

analysis, Dye proposes that the southern part of the ﬁeld

system was primarily developed in the eighteenth and

early nineteenth centuries, with much of the development

throughout the ﬁeld system being a post-contact

phenomenon (Dye 2011b: 29). While it is probable that

people continued to live in and to a limited extent,

develop the LKFS during the early historical era, despite

Dye’s assertions there is no strong archaeological

evidence that signiﬁcant amounts of agricultural

development occurred during the historical era. Ranching

in the LKFS did begin in the nineteenth century, but the

archaeological remains of these activities overlay the

gardening features. Dye concludes that his Bayesian

chronology for LKFS development is sufﬁciently precise

to correlate it with known historical ﬁgures. Yet this is

hardly the case, as he only speciﬁes three people (‘Umi a

Lı¯loa, Kalani‘o¯pu‘u and Kamehameha) who lived within a

time period that spanned some 200–250 years. His

association of these individuals with the ﬁeld system is

hardly novel. His suggestion that developments within the

LKFS represent an effort to create and maintain pig herds

as wealth-assets as opposed to other subsistence- and

surplus-related activities, while intriguing, is not well

substantiated. Dye is correct that we should avoid ad hoc

interpretations and ground our explanations in empirical

data, and future research should test the assumptions and

validity of both good-year and bad-year economic

approaches.

––––––

THOMAS S. DYE

Honolulu, Hawaii. tsd@tsdye.com

The radiocarbon revolution in archaeology is in its third

phase (Bayliss 2009). The initial phase in the 1950s and

1960s saw archaeologists in Hawai‘i use the imprecise

results of the new method to explore the time depth of

Hawaiian prehistory, an area of inquiry that was previously

the sole province of genealogists. The development of

calibration in the 1970s, along with the increasingly

precise results yielded by technical improvements at the

dating laboratories, made it possible for archaeologists to

evaluate and compare local processes of change. The move

away from diffusionism during this second phase of the

revolution led in Hawai‘i to a concern with how the state

institutions of traditional society developed in situ,a

project that has borne considerable fruit (e.g. Hommon

2013; Kirch 2010b). The third phase, which was

introduced by the development of Bayesian calibration,

bridged the gap between the chronological data yielded by

Archaeology in Oceania 21

radiocarbon dating and other sources of relative and

absolute chronological information typically available to

archaeologists, most notably the stratigraphic information

recovered by careful archaeological excavation. Combining

these disparate sources of information makes it possible

for archaeologists to achieve undreamt of levels of

chronological precision. In the United Kingdom, where

application of Bayesian methods is most advanced,

archaeologists have radically revised their understanding

of the early Neolithic period.

Detailed chronologies of change, which at some early

Neolithic sites approach generational timescales, have

shown archaeologists there how “a very different kind of

history can emerge” (Whittle et al. 2011: 1), one that

brings into focus “the social context in which agency,

change and the choices of individuals and communities

occurred” (Whittle et al. 2011: 4).

The ﬁne-grained chronologies of the development of

agricultural infrastructure and construction of family-sized

temples in the Leeward Kohala Field System (LKFS) – the

centrepiece of my paper – are products of this third phase

of the radiocarbon revolution. Cochrane’s comment notices

the positive effect that ﬁne-grained chronologies of change

have on interpretation, the only comment to do so. This

might be taken as an indication that the third phase of the

radiocarbon revolution is just beginning to penetrate the

thinking of Hawaiian archaeologists. Another indication is

Ladefoged’s criticism of one of the prior probabilities used

in the Bayesian model for the agricultural infrastructure.

Model-building is central to the Bayesian project because

it provides the structure needed to integrate radiocarbon

with other sources of chronological information. Thus, it is

important to understand that the prior probabilities of the

model are based on speciﬁc pieces of historical and

stratigraphic information and not on ad hoc assumptions.

In fact, the historical basis for modelling the end of ﬁeld

system development in the mid-nineteenth century

includes census data and the archaeological record of

enclosed habitation sites with historical-era artefact

deposits, and is set out in detail in my paper. Ladefoged’s

comment suggests that the prior probability should be

changed, but does not indicate how it should be changed.

To be clear, treating this prior probability as an assumption

that can be altered so that the posterior probability takes

on particular characteristics – for example, so that it does

not “ramp up” (McCoy et al. 2012: 1208–9) in the

nineteenth century – would be antithetical to scientiﬁc

inquiry. Similarly, dropping a “strict adherence” to data is

not an option open to a scientist. One possible way

forward is to marshal historical information on when ﬁeld

system development stopped, revise the model accordingly

and demonstrate the difference it makes. The tools for this

kind of analysis are well developed and freely available. It

is just a matter of whether or not the historical data can be

found. In any event, I certainly agree with the larger point

behind Ladefoged’s criticism, namely that the demise of

the traditional Hawaiian ﬁeld system is ripe for additional

investigation. My ideas on how this might be carried out

are presented in the paper, but other approaches are

certainly possible.

I am extremely grateful for Earle’s claim that my

argument for the importance of pigs in the political

economy of protohistoric Hawai‘i “suggests a

reconciliation between two materialist traditions that have

often been seen in conﬂict”. I believe the key here is

Bell’s cross-cultural concept of wealth-assets, which

makes it possible to distinguish the large pig herds

encountered by Cook and others as wealth-assets, rather

than consumables.

The idea that traditional Hawaiian pig herds were

created and managed as wealth-assets is resisted by Kirch,

who (i) argues that Cook was mistaken about the size of

the pig herds because his observations were made over a

short period of time at the royal centre of Kealakekua,

where he was treated as a god, and (ii) espouses an

approach that treats pigs as “a signiﬁcant form of

consumable wealth”. However, the historical record for

large pig herds is robust. My paper cites three English

sources – Clerke, King and Vancouver – and notes native

testimony of large pig sacriﬁces. Abundant pigs were

reported throughout the group and not just at Kealakekua.

As Cordy notes in his comment, Clerke singled out Kaua‘i

Island for its large pig herds, and Hawaiian tradition

claims that Captain Brown got 400 pigs at O‘ahu

(Kamakau 1992: 169). The surfeit of pigs at Kealakekua

was not a one-time aberration. Abundant pigs were noted

by Cook and his crew at Kaua‘i in 1778 and by Vancouver

when he visited the islands 15 years later in 1793. The

Brown incident at O‘ahu took place in 1794. Cook

participated in a wide range of pig exchanges, not just

ones that might be interpreted as gifts to a god, and was

well aware of differences among them (Dye 2011a). He

risked mutiny to create conditions in which he hoped to

barter and not exchange gifts. Once he achieved this, he

found that Hawaiians drove a keen bargain and were

willing to take their pigs and other foods home “rather

than despose of them for less than they demand”

(Beaglehole 1967: 483). This is not the behaviour one

expects of a person making offerings to a god.

It is important to understand that treating pigs as

consumables limits an analysis of their role in traditional

Hawaiian society. This can be shown by comparing

Kirch’s concept of “consumable wealth” with Bell’s

concept of wealth-assets. Bell calls the four criteria for the

cross-cultural identiﬁcation of wealth-assets the growth

criterion, the consumption criterion, the marginal value

criterion and the indeﬁnite life criterion. A valued item that

fails to meet one or more of these criteria is a consumable.

Kirch’s comment mentions the growth criterion –

consonant with his longstanding interest in the relationship

of pig production to population size – and the

consumption criterion when he characterises the

distribution of pigs in traditional Hawai‘i. However, his

comment stops here, making it clear that his concept of

“consumable wealth” does not include the marginal value

or indeﬁnite life criteria. Restricting the deﬁnition of

22 Wealth in old Hawai‘i

wealth in this way conﬂates consumables with

wealth-assets, and makes it impossible to see the

constructive ways in which Hawaiian chiefs created and

managed pig herds as wealth-assets. In addition, pigs and

the agricultural ﬁelds that supported animal husbandry are

both visible in the archaeological record, unlike other

traditional Hawaiian wealth-assets such as feather cloaks

and large canoes, which are mostly invisible to

archaeologists. Furthermore, exclusive concern with the

growth and consumption criteria draws a simple opposition

between commoners as producers and chiefs as

consumers. The political dynamic generated by rivalry for

status among the chieﬂy nobility, indicated so clearly in

the comparative structural analysis of Polynesian societies

(Goldman 1970), is lost in this view. It is difﬁcult to

imagine that there is something to gain by ignoring this

part of Hawaiian history.

Conﬂating consumables and wealth-assets also places

restrictions on archaeology’s contribution to regional

comparisons. A structural model of social stratiﬁcation in

Oceania (Friedman 1981) contrasts West Polynesian

societies – in which prestige-good systems were

characterised by generalised marriage exchange, an elite

monopoly on foreign prestige goods, a tendency to

establish wife-taking and wife-giving relationships

between senior and junior lines of the conical clan, and

an asymmetric political dualism that opposed sacred

and secular chiefs, land and sea people, men and women,

and centre and periphery – with Eastern Polynesian

societies where long distances between island groups

greatly increased the difﬁculty of establishing and

maintaining supplies of foreign prestige goods and

removed the material basis on which a prestige-goods

system could develop (Hage & Harary 1996: 116–24). In

this view, the decline of prestige-goods systems in East

Polynesia led to increased competition among elites that

was expressed in warfare and in feasting, which had its

basis in agricultural intensiﬁcation. Here, the theoretical

elaboration of pig herds as wealth-assets and the

connection between sweet potato agriculture and animal

husbandry makes it possible to recognise that the late

intensiﬁcation of infrastructure and construction of

family-sized temples within the LKFS indicates the pace

at which this characteristic feature of East Polynesian

social stratiﬁcation developed in Hawai‘i.

It is too late in Hawai‘i to carry out the detailed

ethnographic description that served as the basis for

identifying prestige-goods systems in Tonga (e.g. Bott

1982; Kaeppler 1978), or that led to the development of

conjectural histories of agricultural production and pig

husbandry as a way of explaining variations in warfare,

leadership, social structure, male–female relationships and

ceremonial exchange among highland Papua New Guinea

societies (Feil 1987). However, the ﬁne-grained

chronologies of change yielded by the Bayesian

calibrations, coupled with a suitable theoretical framework,

open the door to (i) recognising the constructive role

played by chiefs in the development of social stratiﬁcation

during good years and (ii) regional comparisons of

developmental histories.

Like our colleagues in the United Kingdom, the third

phase of the radiocarbon revolution has put Hawaiian

archaeologists in a position to write a “very different kind

of history”. I am conﬁdent that ﬁne-grained chronologies

of change produced by Bayesian analyses are the bright

future of Hawaiian archaeology, and that interpretations

based on them will someday replace the equiﬁnality-prone

arguments based on synchronic data that characterize the

ﬁeld today.

NOTES

1. The Hawaiian word maka‘a¯inana is typically glossed as

“commoner” in that word’s general sense of “one of the

common people . . . below the rank of a peer” (OED). A more

speciﬁc sense of “commoner” is “one who has a joint right in

common lands; one who enjoys a right of common” (OED).

Elsewhere, I have tried to show that Hawaiian maka‘a¯ianana

have an historically variable set of rights with respect to

common lands (Dye 2010b).

2. See http://nupepa.org/cgi-bin/nupepa (accessed 19 March

2014).

REFERENCES

Abad, C.K.C. 2000. The Evolution of Hawaiian Socio-Political

Complexity: An Analysis of Hawaiian Oral Traditions.

Unpublished PhD thesis, University of Hawai‘i at Ma¯noa,

Honolulu, HI.

Adams, J. and Athens, J.S. 1994. Archaeological Inventory

Survey, Upland Portions of Kukuipahu and Awalua, North

Kohala, Island of Hawaii. Prepared for Chalon International

of Hawaii, International Archaeological Research Institute,

Inc., Honolulu, HI.

Allen, J. 2000. Introduction. In Bishop Museum (ed.), Four Sites

in Upland Ka¯ne‘Ohe: Supplemental Inventory Survey for

Interstate Route H-3, pp. 1–30. State Historic Preservation

Division Library, Kapolei, Hawai‘i. On ﬁle.

Allen, M.S. 1996. Style and function in East-Polynesian

ﬁshhooks. Antiquity 70: 97–116.

Ashelford, J. 1996. The Art of Dress: Clothes and Society,

1500–1914. National Trust, London.

Barrau, J. 1965. L’humide et le sec: An essay on ethnobiological

adaptation to contrastive environments in the Indo-Paciﬁc

area. Journal of the Polynesian Society 74: 329–346.

Barrère, D. 1983. Notes on the lands of Waimea and Kawaihae.

In J.T. Clark and P.V. Kirch (eds), Archaeological

Investigations of the Mudlane–Waimea–Kawaihae Road

Corridor, Island of Hawai‘i: An Interdisciplinary Study of an

Environmental Transect, pp. 25–35. Departmental Report

Series 83-1. Anthropology Department, B. P. Bishop Museum,

Honolulu, HI.

Bayliss, A. 2009. Rolling out revolution: Using radiocarbon

dating in archaeology. Radiocarbon 51: 123–147.

Bayman, J. and Dye, T. 2013. Hawaii’s Past in a World of

Paciﬁc Islands. The SAA Press, Washington, DC.

Beaglehole, J.C. (ed.). 1967. The Journals of Captain James

Cook on His Voyages of Discovery. The Voyage of the

Resolution and Discovery, 1776–1780. Vol. 3. Cambridge

University Press, for the Hakluyt Society, Cambridge,

UK.

Archaeology in Oceania 23

Bell, D. 2004. Wealth and Power: Survival in a Time of Global

Accumulation. Alta Mira, Walnut Creek, CA.

Bott, E. 1982. Tongan Society at the Time of Captain Cook’s

Visits: Discussions with Her Majesty Queen Sa¯lote Tupou.

Memoirs of the Polynesian Society 44. Polynesian Society,

Wellington.

Buck, C.E., Cavanagh, W.G. and Litton, C.D. 1996. Bayesian

Approach to Interpreting Archaeological Data. John Wiley &

Sons, Chichester, UK.

Cameron, C.M. 1991. Structure abandonment in villages.

Archaeological Method and Theory 3: 155–194.

Cameron, C.M. and Tomka, S.A. (eds) 1993. Abandonment of

Settlements and Regions: Ethnoarchaeological and

Archaeological Approaches. Cambridge University Press,

Cambridge, UK.

Chadwick, O., Vitousek, P. and Hotchkiss, S. 2013. Farming the

Rock: A Biogeochemical Perspective on Intensive Agriculture

in Polynesia. Paper presented at the 78th Annual Meeting of

the Society for American Archaeology.

Clark, J.T. 1988. Paleodemography in leeward Hawaii.

Archaeology in Oceania 23: 22–30.

Clark, M.R. and Rechtman, R.B. 2004. An Archaeological

Inventory Survey of TMK:3–5–7–02:12, Kamano, Ma¯hukona

1st and 2nd, and Hihiu Ahupua‘a, North Kohala District,

Island of Hawai‘i. Prepared for Kaupulehu Land, LLC,

Rechtman Consulting, Kea‘au, HI.

Clark, M.R., Bury, B.G. and Rechtman, R.B. 2010. An

Archaeological Inventory Survey of TMK:3–5–6–01:108,

Puakea and Kukuipahu Ahupua‘a, North Kohala District,

Island of Hawai‘i. Prepared for Wayne and Linda Thomas,

Rechtman Consulting, Hilo, HI.

Cook, J. and King, J. 1784. A Voyage to the Paciﬁc Ocean.G.

Nicol, London.

Cordy, R. 1972. The effects of European contact on Hawaiian

agricultural systems – 1778–1819. Ethnohistory 19 (4):

393–418.

Cordy, R. 1974a. Cultural adaptation and evolution in Hawaii: A

suggested new sequence. Journal of the Polynesian Society

83: 180–191.

Cordy, R. 1974b. Complex-rank cultural systems in the

Hawaiian Islands: Suggested explanations for their origin.

Archaeology & Physical Anthropology in Oceania 9: 89–109.

Cordy, R. 1978. A Study of Prehistoric Social Change: The

Development of Complex Societies in the Hawaiian Islands.

Unpublished PhD thesis, University of Hawai‘i at Ma¯noa,

Honolulu, HI.

Cordy, R. 1981. A Study of Prehistoric Social Change: The

Development of Complex Societies in the Hawaiian Islands.

Academic Press, New York.

Cordy, R. 2001. Waianae Valley Ranch Archaeological Survey,

Wai‘Anae Ahupua‘A, Moku O Wai‘Anae, O‘Ahu (TMK:

8-5-06:1). State Historic Preservation Division Manuscript.

On ﬁle, Hawai‘i Historic Preservation Division Library,

Kapolei, Hawai‘i.

Cordy, R. 2004. Considering archaeological indicators of the rise

of appointed chiefs and the feudal-land system in the

Hawaiian Islands. Hawaiian Archaeology 9: 1–24.

Cordy, R. 2007. Reconstructing Hawaiian population at

European contact – Three regional case studies. In P. Kirch

and J.-L. Rallu (eds), The Growth and Collapse of Paciﬁc

Island Societies: Archaeological and Demographic

Perspectives, pp. 108–128. University of Hawaii Press,

Honolulu, HI.

Cordy, R. 2012. Book review: How chiefs became kings: Divine

kingship and the rise of archaic states in ancient Hawai‘i,by

Patrick Kirch. The Historian 74 (2): 383–384.

Cordy, R. 2013. Book review: A shark coming inland is my

chief: The island civilization of ancient Hawai‘i,by

Patrick Kirch. Cambridge Archaeological Journal 23 (3):

553–556.

Dixon, B., Gosser, D. and Williams, S. 2008. Traditional

Hawaiian men’s houses and their socio-political context in

Lualulalei, Leeward West O‘ahu, Hawai‘i. Journal of the

Polynesian Society 147: 267–296.

Dixon, B., Conte, P.J., Nagahara, V. and Hodgins, W.K. 2000.

Kahikinui Mauka: Archaeological Research in the Lowland

Dry Forest of Leeward East Maui. Prepared for Department of

Hawaiian Home Lands, Historic Preservation Division,

Department of Land and Natural Resources, Kapolei, HI.

Djindjian, F. 2001. Artefact analysis. In Z. Stancˇicˇ and T.

Veljanovski (eds), Computing Archaeology for Understanding

the Past, pp. 41–52. BAR International Series 931,

Archaeopress, Oxford.

Dye, T.S. 2000. Effects of

C sample selection in archaeology:

An example from Hawai‘i. Radiocarbon 42: 203–217.

Dye, T.S. 2004. How to ﬁx the inventory survey rule. Hawaiian

Archaeology 9: 123–132.

Dye, T.S. 2010a. Social transformation in old Hawai‘i: A

bottom-up approach. American Antiquity 75: 727–741.

Dye, T.S. 2010b. Traditional Hawaiian surface architecture:

Absolute and relative dating. In T.S. Dye (ed.), Research

Designs for Hawaiian Archaeology: Agriculture, Architecture,

Methodology, pp. 93–155. Special Publications 3, Society for

Hawaiian Archaeology, Honolulu, HI.

Dye, T.S. 2011a. A model-based age estimate for Polynesian

colonization of Hawai‘i. Archaeology in Oceania 46:

130–138.

Dye, T.S. 2011b. The tempo of change in the Leeward Kohala

Field System, Hawai‘i Island. Rapa Nui Journal 25:

21–30.

Dye, T.S. 2011c. Gift exchange and interpretations of Captain

Cook in the traditional kingdoms of the Hawaiian Islands.

Journal of Paciﬁc History 46: 275–292.

Dye, T.S. 2012. Hawaiian temples and Bayesian chronology.

Antiquity 86: 1202–1206.

Dye, T.S. and Komori, E.K. 1992. A pre-censal population

history of Hawai‘i. New Zealand Journal of Archaeology 14:

113–128.

Dye, T.S. and Pantaleo, J. 2010. Age of the O18 site, Hawai‘i.

Archaeology in Oceania 45: 113–119.

Earle, T. 1973. Control Hierarchies in the Traditional Irrigation

Economies of Halelea District, Kauai, Hawaii. Unpublished

PhD thesis, University of Michigan, Ann Arbor.

Earle, T. 1978. Economic and Social Organization of A Complex

Chiefdom: The Halelea District, Kaua‘i, Hawaii.

Anthropological Papers 63, Museum of Anthropology,

University of Michigan, Ann Arbor, MI.

Earle, T. 1980. Prehistoric irrigation in the Hawaiian Islands: An

evaluation of evolutionary signiﬁcance. Archaeology and

Physical Anthropology in Oceania XV: 1–28.

Earle, T. 2012. Taro irrigation and primary state formation in

Hawai’i. In M. Spriggs, D. Addison and P. Matthews (eds),

Irrigated Taro (Colocasia esculenta) in the Indo-Paciﬁc, pp.

95–114. National Museum of Ethnology, Osaka.

Emory, K.P. and Sinoto, Y.H. 1969. Age of the Sites in the South

Point Area, Ka‘u, Hawaii. Paciﬁc Anthropological Records 8,

Anthropology Department, Bernice P. Bishop Museum,

Honolulu, HI.

Feil, D.K. 1987. The Evolution of Highland Papua New Guinea

Societies. Cambridge University Press, Cambridge, UK.

Field, J. 2004. Environmental and climatic considerations: A

hypothesis for conﬂict and the emergence of social

24 Wealth in old Hawai‘i

complexity in Fijian prehistory. Journal of Anthropological

Archaeology 23: 79–99.

Field, J.S., Ladefoged, T.N. and Kirch, P.V. 2011. Household

expansion linked to agricultural intensiﬁcation during

emergence of Hawaiian archaic states. Proceedings of the

National Academy of Sciences USA 108: 7327–7332.

Firth, R. 1959. Social Change in Tikopia: Re-study of a

Polynesian Community after a Generation. George Allen &

Unwin, London.

Fornander, A. 1916–1919. Fornander Collection of Hawaiian

Antiquities and Folklore. Memoirs of the B. P. Bishop

Museum 4–6, Bishop Museum Press, Honolulu, HI.

Fox, A. 1976. Prehistoric Maori Fortiﬁcations in the North

Island of New Zealand. Longman Paul, Auckland.

Friedman, J. 1974. Marxism, structuralism and vulgar

materialism. Man (n.s.) 9: 444–469.

Friedman, J. 1981. Notes on structure and history in Oceania.

Folk 23: 275–295.

Geertz, C. 1963. Agricultural Involution; The Process of

Ecological Change in Indonesia. Association for Asian

Studies Monographs and Papers 11, University of California

Press, Berkeley, CA.

Goldman, I. 1970. Ancient Polynesian Society. University of

Chicago Press, Chicago.

Graves, D.K. 1992. Archaeological Inventory Survey, Kohala

Makai-I Project Area, Land of Waika, North Kohala District,

Island of Hawaii. Prepared for Belt, Collins & Associates,

Paul H. Rosendahl, Ph.D., Inc., Hilo, HI.

Graves, D.K. and Franklin, L.J. 1998. Archaeological Inventory

Survey, Kahua Makai/Kahua Shores Coastal Parcels, Lands of

Kahua 1 and 2 and Waika, North Kohala District, Island of

Hawai‘i (TMK:3–5–9–01:7, 8). Prepared for Gentry Hawaii,

Ltd. Paul H. Rosendahl, Ph.D., Inc., Hilo, HI.

Graves, M.W., Cachola-Abad, C.K. and Ladefoged, T.N. 2010.

The evolutionary ecology of Hawaiian political complexity.

In P.V. Kirch (ed.), Roots of Conﬂict: Soils, Agriculture,

and Sociopolitical Complexity in Ancient Hawai‘i, pp.

135–162. School for Advanced Research Press, Santa

Fe, NM.

Hage, P. and Harary, F. 1996. Island Networks: Communication,

Kinship, and Classiﬁcation Structures in Oceania. Cambridge

University Press, Cambridge, UK.

Hammatt, H.H. and Borthwick, D.F. 1986. Archaeological

Survey and Excavations at Kohala Ranch, North Kohala,

Hawaii Island. Prepared for Kohala Ranch, Cultural Surveys

Hawaii, Kailua, HI.

Handy, E.S.C. 1940. The Hawaiian Planter. Bernice P. Bishop

Museum Bulletin 161, Bishop Museum Press, Honolulu, HI.

Handy, E.S.C. and Handy, E.G. 1972. Native Planters in Old

Hawaii: Their Life, Lore, and Environment. Bernice P.

Bishop Museum Bulletin 233, Bishop Museum Press,

Honolulu, HI.

Handy, E.S.C. and Pukui, M.K. 1972. The Polynesian Family

System in Ka-‘u, Hawai‘i. Charles E. Tuttle, Tokyo.

Hartshorn, A.S., Chadwick, O.A., Vitousek, P.M. and Kirch, P.V.

2006. Prehistoric agricultural depletion of soil nutrients in

Hawai‘i. Proceedings of the National Academy of Sciences

USA 103: 11092–11097.

Hommon, R. 1972. Hawaiian Cultural Systems and

Archaeological Site Patterns. Unpublished M.A. thesis,

University of Arizona, Tuscon, AZ.

Hommon, R.J. 1976. The Formation of Primitive States in

Pre-contact Hawaii. Unpublished PhD thesis, University of

Arizona, Tucson, AZ.

Hommon, R.J. 1986. Social evolution in ancient Hawaii. In P.V.

Kirch (ed.), Island Societies: Archaeological Approaches to

Evolution and Transformation, pp. 55–68. Cambridge

University Press, Cambridge, UK.

Hommon, R.J. 2010. Watershed: Testing the limited land

hypothesis. In T.S. Dye (ed.), Research Designs for Hawaiian

Archaeology: Agriculture, Architecture, Methodology, pp.

1–92. Special Publications 3, Society for Hawaiian

Archaeology, Honolulu, HI.

Hommon, R.J. 2013. The Ancient Hawaiian State: Origins of a

Political Society. Oxford University Press, Oxford.

Kaeppler, A.L. 1978. Exchange patterns in goods and spouses:

Fiji, Tonga and Samoa. Mankind 11: 246–252.

Kaeppler, A.L. and de Rooij, W. 2010. Hawaiian Featherwork:

Catalogue Raisonné of Pre-1900 Feathered-God Images,

Cloaks, Capes, Helmets. Feymedia Verlagsgesellschaft,

Düsseldorf.

Kagawa, A.K. and Vitousek, P.M. 2012. The Ahupua‘a of

Puanui: A resource for understanding Hawaiian rain-fed

agriculture. Paciﬁc Science 66: 161–172.

Kamakau, S.M. 1964. Ka Po’e Kahiko: The People of Old.

Trans. M. K. Pukui. Bernice P. Bishop Museum Special

Publication No. 51. Bishop Museum Press, Honolulu, HI.

Kamakau, S.M. 1992. Ruling Chiefs of Hawaii. Revised,

Kamehameha Schools Press, Honolulu, HI.

Kirch, P.V. 1984. The Evolution of the Polynesian Chiefdoms.

Cambridge University Press, Cambridge, UK.

Kirch, P.V. 1994. The Wet and the Dry: Irrigation and

Agricultural Intensiﬁcation in Polynesia. University of

Chicago Press, Chicago.

Kirch, P.V. 2001. Polynesian feasting in ethnohistoric,

ethnographic, and archaeological contexts: A comparison of

three societies. In M. Dietler and B. Hayden (eds), Feasts:

Archaeological and Ethnographic Perspectives on Food,

Politics, and Power, pp. 168–184. Smithsonian Institution

Press, Washington, DC.

Kirch, P.V. 2007. Like shoals of ﬁsh: Archaeology and population

in pre-contact Hawai’i. In P. Kirch and J.-L. Rallu (eds), The

Growth and Collapse of Paciﬁc Island Societies:

Archaeological and Demographic Perspectives, pp. 52–69.

University of Hawaii Press, Honolulu, HI.

Kirch, P.V. 2010a. How Chiefs Became Kings: Divine Kingship

and the Rise of Archaic States in Ancient Hawai‘i. University

of California Press, Berkeley, CA.

Kirch, P.V. (ed.). 2010b. Roots of Conﬂict: Soils, Agriculture, and

Sociopolitical Complexity in Ancient Hawai‘i. School for

Advanced Research Press, Santa Fe, NM.

Kirch, P.V. 2012. A Shark Coming Inland Is My Chief: The

Island Civilization of Ancient Hawai’i. University of

California Press, Berkeley.

Kirch, P.V. and Sahlins, M. 1992. Anahulu: The Anthropology of

History in the Kingdom of Hawaii, 2 vols. University of

Chicago Press, Chicago.

Kirch, P.V. and Zimmerer, K.S. 2010. Dynamically coupled

human and natural systems: Hawai‘i as a model system.

In P.V. Kirch (ed.), Roots of Conﬂict: Soils, Agriculture,

and Sociopolitical Complexity in Ancient Hawai‘i,

pp. 3–30. School for Advanced Research Press, Santa Fe,

NM.

Kirch, P.V., Asner, G., Chadwick, O.A., Field, J., Ladefoged, T.,

Lee, C., Puleston, C., Tuljapurkar, S. and Vitousek, P.M.

2012. Building and testing models of long-term agricultural

intensiﬁcation and population dynamics: A case study from

the Leeward Kohala Field System, Hawai‘i. Ecological

Modelling 227: 18–28.

Kolb, M. 1991. Social Power, Chieﬂy Authority, and Ceremonial

Architecture in an Island Polity, Maui, Hawaii. Unpublished

PhD thesis, UCLA, Los Angeles, CA.

Archaeology in Oceania 25

Kolb, M. 1999. Staple ﬁnance, ritual pig sacriﬁce, and

ideological power in ancient Hawai’i. In E. Bacus and L.

Lucero (eds), Complex Polities in the Ancient Tropical World.

Archaeological Papers of the American Anthropological

Association, No. 9, pp. 89–107.

Kolb, M.J. 2006. The origins of monumental architecture in

ancient Hawai‘i. Current Anthropology 47: 657–665.

Kolb, M., Conte, P. and Cordy, R. 1997. Kula: the archaeology

of upcountry Maui in Waiohuli and Ke¯o¯kea: An

archaeological and historical settlement survey in the

Kingdom of Maui life expectancy and surplus production of

dynamic pre-contact territories in leeward Kohala, Hawai’i.

Manuscript on ﬁle, Hawai‘i State Historic Preservation

Division Library, Kapolei, Hawai‘i.

Kurashima, N. and Kirch, P.V. 2011. Geospatial modeling of

pre-contact Hawaiian production systems on Molok‘i Island,

Hawaiian Islands. Journal of Archaeological Science 38:

3662–3674.

Ladefoged, T.N. and Graves, M.W. 2008. Variable development

of dryland agriculture in Hawai‘i: A ﬁne-grained chronology

from the Kohala Field System, Hawai‘i Island. Current

Anthropology 49: 771–802.

Ladefoged, T.N. and Graves, M.W. 2010. The Leeward Kohala

Field System. In P.V. Kirch (ed.), Roots of Conﬂict: Soils,

Agriculture, and Sociopolitical Complexity in Ancient

Hawai‘i, pp. 89–110. School for Advanced Research Press,

Santa Fe, NM.

Ladefoged, T.N., Lee, C.T. and Graves, M.W. 2008. Modeling

life expectancy and surplus production of dynamic pre-contact

territories in leeward Kohala, Hawai’i. Journal of

Anthropological Archaeology 27: 93–110.

Ladefoged, T.N., Kirch, P.V., Chadwick, O.A., Gon, III S.M.,

Hartshorn, A.S. and Hotchkiss, S.C. 2010. Hawaiian

agro-ecosystems and their spatial distribution. In P.V. Kirch

(ed.), Roots of Conﬂict: Soils, Agriculture, and Sociopolitical

Complexity in Ancient Hawai‘i, pp. 45–63. School for

Advanced Research Press, Santa Fe, NM.

Law, R.G. and Green, R.C. 1972. An economic interpretation of

Taniwha P¯a, Lower Waikato, New Zealand (N52/1). Mankind

8: 255–269.

Lee, C. and Tuljapurkar, S. 2008. Population and prehistory I.

Food-dependent population growth in constant environments.

Theoretical Population Biology 73: 473–483.

Lee, C. and Tuljapurkar, S. 2010. Quantitative, dynamic models

to integrate environment, population, and society. In P.V.

Kirch (ed.), Roots of Conﬂict: Soils, Agriculture, and

Sociopolitical Complexity in Ancient Hawai‘i, pp. 111–133.

School for Advanced Research Press, Santa Fe, NM.

Lee, C., Tuljapurkar, S. and Vitousek, P.M. 2006. Risky business:

Temporal and spatial variation in preindustrial dryland

agriculture. Human Ecology 34: 739–763.

Lee, R.D. 1986. Malthus and boserup: A dynamic synthesis. In

D. Coleman and R.S. Schoﬁeld (eds), The State of Population

Theory: Forward from Malthus, pp. 96–103. Basil Blackwell,

Oxford.

Lincoln, N. 2013. Kona Field System research. Personal

communication, February.

Linnekin, J. 1988. Who made the feather cloaks? A problem in

Hawaiian gender relations. Journal of the Polynesian Society

97: 265–280.

Linnekin, J. 1990. Sacred Queens and Women of Consequence.

University of Michigan Press, Ann Arbor, MI.

Lipo, C.P., Hunt, T.L. and Hundtoft, B. 2010. Stylistic variability

of stemmed obsidian tools (mata’a), frequency seriation, and

the scale of social interaction on Rapa Nui (Easter Island).

Journal of Archaeological Science 37: 2551–2561.

Lockwood, C. 2009. Pigs, Dryland Agriculture and Social

Complexity in Precontact Hawai‘i: Assessing Surplus

Production through Landscape Geochemistry. Unpublished

PhD Dissertation, Department of Anthropology, University of

Washington, Washington, DC.

Loubser, J.H.N. and Rechtman, R.B. 2007. Archaeological Data

Recovery Investigation at SIHP Sites 2492, 16177, 16122,

2494, 16129, 2485, and 16154, within Site Complex

50–10–05–4157 (TMK: 3–5–9–017:001–007), Kahua¯ 2 and

Waika¯Ahupua‘a, North Kohala District, Island of Hawai‘i.

Prepared for Kohala Kai LLC, Rechtman Consulting, Kea‘au,

HI.

Macpherson, C.B. 1985. The Rise and Fall of Economic Justice

and Other Papers. Oxford University Press, Oxford.

Malo, D. 1951. Hawaiian Antiquities. Bernice P. Bishop Museum

Special Publication 2, Bishop Museum Press, Honolulu, HI.

Malo, D. 1996. Ka Mo‘Olelo Hawai‘i: Hawaiian Traditions.

First People’s Productions, Honolulu, HI.

McCoy, M.D., Ladefoged, T.N., Graves, M.W. and Stephen, J.W.

2011. Strategies for constructing religious authority in ancient

Hawai‘i. Antiquity 85: 927–941.

McCoy, M.D., Ladefoged, T.N., Bickler, S.H., Stephen, J.W. and

Graves, M.W. 2012. The value of an “eclectic and pragmatic”

approach to chronology building. Antiquity 86: 1206–1209.

Menzies, A. 1920. Hawaii Nei 128 Years Ago. W. F. Wilson,

Honolulu, HI.

Mesoudi, A. 2008. Foresight in cultural evolution. Biology and

Philosophy 23: 243–255.

Meyer, M., Ladefoged, T.N. and Vitousek, P.M. 2007. Soil

phosphorous and agricultural development in the Leeward

Kohala Field System. Paciﬁc Science 61: 347–353.

Nakkim, L.B. 1970. Possible food storage Pits in Hawaii: A

preliminary report on the results of a surface survey at Hana,

Maui, Hawaii. New Zealand Archaeological Association

Newsletter 13: 123–127.

Ormerod, P. 2005. Why Most Things Fail. Faber and Faber,

London.

Pukui, M.K., Hartig, E.W. and Lee, C.A. 1972. Na¯na¯i ke Kumu

(Look to the Source), 2 vols. Hui Ha¯nai (Queen Lili‘uokalani

Children’s Center), Honolulu, HI.

Puleston, C.O. and Tuljapurkar, S. 2008. Population and

prehistory II. Space-limited human population in constant

environments. Theoretical Population Biology 74: 147–160.

Ribeiro, A. 1984. Dress in Eighteenth-Century Europe,

1715–1789. Batsford, London.

Rick, J.W. 1987. Dates as data: An examination of the Peruvian

preceramic radiocarbon record. American Antiquity 52:

55–73.

Rosendahl, P.H. 1994. Aboriginal Hawaiian structural remains

and settlement patterns in the upland agricultural zone at

Lapakahi, Island of Hawai‘i. Hawaiian Archaeology 3:

14–70.

Sahlins, M. 1958. Social Stratiﬁcation in Polynesia. American

Ethnological Society Monograph. University of Washington

Press, Seattle, WA.

Sahlins, M.D. 1992. Historical ethnography. In P.V. Kirch and

M.D. Sahlins (eds), Anahulu: The Anthropology of History in

the Kingdom of Hawaii. Volume 1: Historical Ethnography.

University of Chicago Press, Chicago.

Sahlins, M.D. 1995. How “Natives” Think: About Captain Cook,

for Example. University of Chicago Press, Chicago.

Smith, A. and Gazin-Schwartz, A. (eds) 2008. Landscapes of

Clearance: Archaeological and Anthropological Perspectives.

Left Coast Press, Walnut Creek, CA.

Spriggs, M. 1985. Why irrigation matters in Paciﬁc prehistory.

Journal of Paciﬁc History 20: 23–41.

26 Wealth in old Hawai‘i

Spriggs, M. 1986. Landscape, land use and political

transformation in southern Melanesia. In P. Kirch (ed.), Island

Societies: Archaeological Approaches to Evolution and

Transformation, pp. 6–19. Cambridge University Press,

Cambridge, UK.

Stuiver, M. and Polach, H. 1977. Discussion: Reporting of

data. Radiocarbon 19: 355–363.

Tainter, J. and Cordy, R. 1977. An archaeological analysis of

social ranking and residence groups in prehistoric Hawaii.

World Archaeology 9 (1): 95–112.

Tcherkézoff, S. 2003. On cloth, gifts, and nudity: Regarding

some European misunderstandings during early encounters in

Polynesia. In C. Colchester (ed.), Clothing the Paciﬁc, pp.

51–75. Berg, Oxford.

Tomonari-Tuggle, M.J. 1988. North Kohala: Perception of a

Changing Community, a Cultural Resources Study. Division

of State Parks, Outdoor Recreation, Historic Sites, Department

of Land, and Natural Resources, State of Hawai‘i, Honolulu,

HI.

Tuggle, H.D. 2010. Lady Mondegreen’s hopes and dreams: Three

brief essays on inference in Hawaiian archaeology. In T.S.

Dye (ed.), Research Designs for Hawaiian Archaeology, pp.

157–184. Special Publications 3, Society for Hawaiian

Archaeology, Honolulu, HI.

Tuljapurkar, S., Lee, C. and Figgs, M. 2007. Demography and

food in early Polynesia. In P.V. Kirch and J.-L. Rallu (eds),

The Growth and Collapse of Paciﬁc Island Societies:

Archaeological and Demographic Perspectives, pp. 35–51.

University of Hawai‘i Press, Honolulu, HI.

Valeri, V. 1985a. Kingship and Sacriﬁce: Ritual and Society in

Ancient Hawaii. University of Chicago Press, Chicago.

Valeri, V. 1985b. The conqueror becomes king: A political

analysis of the Hawaiian legend of ‘Umi’. In A. Hooper and J.

Huntsman (eds), Transformations of Polynesian Culture, pp.

79–103. Memoirs of the Polynesian Society 45, Auckland, NZ.

Vancouver, G. 1798. A Voyage of Discovery to the North Paciﬁc

Ocean, and Round the World: In Which the Coast of

North-West America Has Been Carefully Examined and

Accurately Surveyed: Undertaken by His Majesty’s Command,

Principally with A View to Ascertain the Existence of Any

Navigable Communication between the North Paciﬁc and

North Atlantic Oceans, and Performed in the Years 1790,

1791, 1792, 1793, 1794, and 1795, in the Discovery Sloop of

War, and Armed Tender Chatham, Under the Command of

Captain George Vancouver, 3 vols. G.G. and J. Robinson and

J. Edwards, London.

Vitousek, P.M. and Chadwick, O.A. 2013. Pedogenic thresholds

and soil process domains in basalt-derived soils. Ecosystems

16: 1379–1395.

Vitousek, P.M., Chadwick, O.A., Hilley, G., Kirch, P.V. and

Ladefoged, T.N. 2010. Erosion, geological history, and

indigenous agriculture: A tale of two valleys. Ecosystems 13:

782–793.

Vitousek, P.M., Ladefoged, T.N., Kirch, P.V., Hartshorn, A.S.,

Graves, M.W., Hotchkiss, S.C., Tuljapurkar, S. and

Chadwick, O.A. 2004. Soils, agriculture, and society

in Precontact Hawai‘i. Science 304: 1665–1669.

Weiner, A.B. 1992. Inalienable Possessions: The Paradox of

Keeping-While-Giving. University of California Press,

Berkeley, CA.

Whittle, A., Healy, F., Bayliss, A. and Allen, M. 2011.

Gathering Time: Dating the Early Neolithic Enclosures

of Southern Britain and Ireland. Oxbow Books, Oxford.

Willer, D. and Willer, J. 1973. Systematic Empiricism:

Critique of a Pseudoscience. Prentice-Hall, Englewood

Cliffs, NJ.

Wood, J.W. 1998. A theory of preindustrial population dynamics:

Demography, economy, and well-being in Malthusian

systems. Current Anthropology 39: 99–121.

Wulzen, W. and Goodfellow, S.T. 1995. Final Report, Phased

Archaeological Inventory Survey: Phase II – Data Collection,

Chalon International Mahukona Mauka Parcel, Lands of

Kamano, Mahukona 1st and 2nd, Hihiu, and Ka‘oma, North

Kohala District, Island of Hawai‘i, TMK:3–5–7–02:por. 27

and 36. Prepared for Chalon International of Hawaii, Paul H.

Rosendahl, Ph.D., Inc., Hilo, HI.

Yen, D.E. 1974. The Sweet Potato and Oceania. Bernice P.

Bishop Museum Bulletin 236, Bishop Museum Press,

Honolulu, HI.

Archaeology in Oceania 27