Who Detects Ecological Change After Catastrophic Events?

Indigenous Knowledge, Social

Networks, and Situated Practices
Author(s): Matthew Lauer and Jaime Matera
Source: Human Ecology , FEBRUARY 2016, Vol. 44, No. 1 (FEBRUARY 2016), pp. 33-46
Published by: Springer

Stable URL: https://www.jstor.org/stable/24762721

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Springer is collaborating with JSTOR to digitize, preserve and extend access to Human Ecology

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
Hum Ecol (2016)44:33-46
DOI 10.1007/s 10745-016-9811 -3

Who Detects Ecological Change After Catastrophic Events?

Indigenous Knowledge, Social Networks, and Situated Practices
Matthew Lauer1 • Jaime Matera2

Published online: 2 February 2016

£> Springer Science+Business Media New York 2016

Abstract Detecting ecological change is a critical first stepIntroduction

in
the process of local-level adaptation, yet few studies have ex
Over the last several decades, indigenous ecological knowl
plored the factors that predict knowledge acquisition following
catastrophic events. This article empirically assesses individual
edge (IEK) has become a burgeoning field of study in a wide
range of academic disciplines and applied pursuits. Disaster
variation in the ability of Solomon Islanders to detect ecological
change following the alteration of local, shallow-water, marine
experts, for example, have begun to appreciate the importance
environments by a major tsunami. We compare the results
ofof
indigenous knowledge for understanding environmental
marine science surveys with local ecological knowledge of the
hazards, reducing disaster risk, and minimizing vulnerability
benthos. We also examine multiple socioeconomic variables,
(Ellen 2007; Mercer et al. 2007). In sustainable development
and employ social network analysis to measure the influence
and natural resource management, it is argued that when local
of social and expert networks. Results show that villagers with
peoples' perceptions, judgments, and rationales about ecolog
salaried work who are at the intersection of local and global
ical processes are integrated into the development of pro
knowledge were the most adept at detecting tsunami-induced
grams, they tend to be more effective ecologically and socially
(Berkes et al. 2000; Brokensha et al. 1980). This is found
changes to benthic surfaces. Social networks had no statistically
significant influence on villagers' abilities to detect change. particularly
We to be the case in most regions of the world now
argue that these results counter common conceptualizationsexperiencing
of rapid social and ecological change.
indigenous knowledge that emphasize its normative, shared,Advocates of IEK emphasize its richness and site specific
inter-generationally transmitted characteristics rather thanity
its as characteristics that enable local resource users to alter
heterogeneity, emergence, and practical application. Our find
their management strategies as ecological conditions change
ings have implications for theory about the foundations of(Berkes
in et al. 2000; Drew 2005). The interplay between local
knowledge, environmental conditions, and management deci
digenous knowledge research and the design of disaster mitiga
sions plays a fundamental role in long-standing, traditional
tion efforts or resource management programs that incorporate
indigenous ecological knowledge. resource management. IEK-informed practices, in some cases,
support the resilience of socio-ecological systems because
they manage for change rather than specific conservation or
Keywords Tsunami • Disaster mitigation - Ecological
societal outcomes by continually adapting management poli
change Indigenous ecological knowledge Solomon Islands ■
cies (Berkes et al. 2000).
Benthos
A fundamental capacity that enables indigenous people to
adapt is ecological change detection, a form of IEK that has
ISI Matthew Lauer been empirically documented in a variety of contexts
mlauer@mail.sdsu.edu (Alexander et al. 2011; Aswani et al. 2014; Lauer and
Aswani 2010). Until the 1990s, IEK studies rarely evaluated
who detects changing ecosystems. Moreover, the variation,
1 San Diego State University, San Diego, CA, USA distribution and evolution of knowledge within traditional
2 California State University Channel Islands, Camarillo, CA, USA communities as well as the pathways by which knowledge

£) Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 34 Hum Ecol (2016) 44:33-46

flows tended toknowledge

be(Vandebroek and Balick 2012). Knowledge
secondary ero
rath
(for some exceptions see
sion, in particular, is an important Boster
area of research. A signifi
ethnobiological research,
cant body for
of research suggests that as communities are drawnexa
with identifying cross-cultural
into market economies, IEK is displaced by new knowledge r
classification rather
associated with livelihood than the
changes (Reyes-Garcia etal. 2005; dy
tion (e.g., Berlin
Ross 2002;1992). This
Schultes 1994; Zent 2009). gene
For example, Spoon
to the way in which
(2011) showed classic
how the ecological knowledge of the Khumbu IE
knowledge. Typically, knowledg
Sherpa, agro-pastoralists from Nepal, has shifted away from
tive, socially agro-pastoralism towards information and
transmitted values associated sh
trait
or societies, a perspective that
with a growing tourism industry. These changes, however, ha
in the social sciences
were highly variable across (Ellen
the Sherpa population et
and a
More
recently, epistemological
across domains of knowledge. Knowledge erosion appears
aboutapproaches
to vary both to indigenous
at the individual level and across biocultural
mogeneity over
settings. hybridity
Zarger and Stepp (2004) documented that and
even in a c
experiential, and other
highly acculturated setting Tzeltalmodalities
Maya children retain a rich
ing (Ellen et al.
understanding 2000;
of botanical knowledge,Escobar
suggesting that IEK
1993; Ingold 2000; Zent 2009).
may persist among some individuals or segments of a society K
proaches or processural
despite significant cultural and economic perspec
change. Educational
knowledge as settings appear to play an important
dynamic and role in knowledge
situate
emergent, experiential,
transformation. Ohmagari and Berkesand hete
(1997) research among
a result, the Cree women of subarctic
focus of Canada research
revealed how the use of s
shared values, beliefs,
non-local patterns,
language in schools heavily impacted knowledge
of of the to
communities local environment.
understanding
tribution, change, flow,
Very little research, and
however, has empirically em
assessed the
other knowledge.
capacity of local or indigenous people to detect changes after
Our previous large-scale, ecological disruptions.
work in These events may
the Solomperma
practice-based nently
approach to know
alter local ecological conditions, yet little is known
a more comprehensive theoretic
about the extent, accuracy, and rapidity of detection by tradi
ing IEK of thetional local marine
people, and how the related eco
new knowledge propagates.
2009) as well as disaster
Clearly, respons
the type and extent of disturbance will influence
document how knowledge
IEK acquisition:
is slow ecological
not change, such as
only
generationally'shifting
through oral
baselines,' will be detected differently hist
than abrupt,
vised, bodily, non-cognitive
intense ecological disturbances caused by a tsunami. way
cases are overlooked
To explore pathways of by cogniti
knowledge acquisition and its
knowledge. A transmission
situated practice
through communities, researchers in many fields
conceptual space for
have employed these
social network more
analysis, although to date only
ing, it also provides
a handful of studies have a
appliedtheoretica
these techniques to IEK
lematic distinction
(Bodin and Prell 2011; Reyes-Garcia et al. 2013). This is quiteloc
between
knowledge (Agrawal 1995)
surprising considering the explosion of interest in by
network c
practices (Pickering 1995). Tum
studies across the social and natural sciences (Borgatti et al.
"situated messiness" or the "motley" nature of 2009; Wasserman and Faust 1994). One of the most widely
techcnoscientific practice is often overlooked because ofresearched
a concepts in social network studies is centrality,
bias in western thinking towards theory at the expense of the
which refers to the structural prominence or importance of a
actual practice of science. node (e.g., individual, household, organization) in a network.
Propelled by this alternative, practice/processural viewTypically,
of individuals with many non-negative social ties in
knowledge, a raft of recent research has focused on the quan
volving friendship or trust tend to be more effective dissemi
titative measurement of individual variation in knowledge
nators of information (Katona etal. 2011), which is important
(Atran and Medin 2008; Reyes-Garcia et al. 2007). Ethnobi
since studies suggest that occupying central positions within
ological studies in particular have developed increasingly networks
so may result in a higher level of ecological knowledge
phisticated methods and analyses exploring knowledge het (Reyes-Garcia etal. 2013). Many questions remain, however,
erogeneity. Factors such as age, gender, livelihood, residence,
about the extent to which these principles apply to ecological
education, household income, and integration into the market
knowledge in small-scale, traditional communities facing the
economy have been shown to correlate with variation incombined forces of globalization and climate change, or

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 Hum Ecol (2016) 44:33^6 35

which have hamlets (Solomon Islands Government

been subjected 2011). Its inhabitants t
bances such are culturally
as indistinguishable from, and genealogically
earthquakes, t
Here we enmeshed with, people
argue that who reside in adjacent Roviana La
concept
dynamic, goon. In both lagoons, Roviana,
active, and an Austronesian language
heterog
and processes
unique toprovides a
the region, is the local vernacular. Solomon Islands mor
theoretical foundation from
Pijin, the lingua franca of the Solomon Islands, is also widely
Islanders spoken. As in most of the Solomonthe
perceived Islands, Vonavona com
ecolo
We examine the
munities rely onfactors
subsistence fishing and horticulture forthat
their i
knowledge, main livelihoods even though extensive social and
including thecultural eff
2007, an 8.1 change has occurred over the past several centuries. Subsis
magnitude earth
30 km off the coast of the western Solomon Islands. Thetence fishing continues to dominate village life, and marine
quake generated a large tsunami that decimated seaside vil
resources provide most of the protein in the local diet. Many
lages and changed marine ecological conditions. In 2006, our
households also engage intermittently with the cash economy,
research team conducted marine science (MS) surveys
undertaking small-scale commercial activities such as
documenting different habitat types and benthic surfaces shell-diving,
in copra-production or the marketing of fish, shell
Vonavona Lagoon, a large lagoon on the western coast offish, fruits or root crops. Logging operations have also prolif
New Georgia Island. After the tsunami we replicated theerated, and provide employment for many islanders.
2006 survey at two equally spaced time intervals (2008 and For this research, social science fieldwork was conducted
2010) to document any changes to the benthos. in 2010 among villagers in three communities located in
southern Vonavona: Kinda, Saika, and Kinamara. The three
Previous analyses comparing IEK and the marine surveys
villages have approximately 110 households with a total pop
indicate that the tsunami changed benthic habitats and that
local Solomon Islanders detected the changes to their seaulation of 500 individuals. Several decades ago these commu
scape. (Aswani and Lauer 2014). This previous research, nities lived together on a small island in the center of the
Lagoon called Repi. In the mid-1990s, however, the site was
however, assessed the abilities of villagers in focus groups.
Below, we explore in more detail the capacity of individualabandoned because it lacked a reliable source of fresh water,
and the population was growing rapidly, causing overcrowd
villagers, rather than focus groups, to detect environmental
changes by considering how specific variables identified in ing and other issues.
the IEK literature (e.g., age, gender, occupation, education, Prior to the 2007 tsunami, our study team conducted MS
etc.) may influence the ability to monitor local ecologies. surveys
In of an inner lagoon reef system that extends off the
addition, we employ social network analysis to ascertainsoutheastern edge of Repi Island. We were studying the region
whether the perpetuation and transmission of knowledgeasispart of a marine resource management initiative that began
shaped by positions within social and expert networks. in 2000 (Aswani et al. 2004). The 10-year initiative culminat
ed in the creation of a network of marine protected areas
(MPAs), including eight MPAs in Vonavona Lagoon. To
Methods achieve this we (and others from the research team including
local counterparts) conducted numerous participatory work
Study Site shops in Vonavona Lagoon to assist local communities in the
establishment and maintenance of their management program.
The Solomon Islands make up the third largest archipelago in The process also involved: a) the development of a local
the Pacific, with six main islands and nearly 1000 smaller non-governmental organization called Tiola Conservation
islets. Vonavona Lagoon lies off the western coast of New Foundation, b) the integration of the management initiatives
Georgia, and is bounded by two smaller islands, Parara to into the regional and national governments' costal manage
the west and Kohinggo to the east (Fig. 1). The 30 km-long ment plans, c) legally codifying the management initiatives in
lagoon contains numerous protected bays, barrier islands, government statutes and legislation, and d) training local
pools, coral reefs, intertidal flats, and passages, and is world counterparts in marine science, anthropology, and the use of
renowned for its marine biodiversity and striking natural beau Geographic Information Systems (GIS). This long-term col
ty (Green et al. 2006). Large portions of the lagoon are shal laboration with local villagers, governmental officials, and
low, (<20 m) and contain a multitude of marine habitats, in Solomon Islander counterparts helped ensure that Vonavona
cluding seagrass meadows, mangroves, freshwater swamps, villagers involved in our study were confident that it would
river estuaries, sand channels, shallow coral reefs, silt-laden benefit their community and region.
embayments, and reef drops. The Repi reefs were initially selected for the benthic sur
Vonavona Lagoon is home to approximately 5,500 people veys because of the diversity of its inner lagoon habitats in
who reside in several main villages and a half-dozen smaller cluding seagrass, coral reefs, sand flats, pools, and silt.

<?} Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
36 Hum Ecol (2016) 44:33-46

Fig. 1 Solomons Islands with inset o

Moreover, only
Vonavona a biophysic
villagers re
for the bulk of and people's
their protein,ide
and
(Aswani 1998). of
The a named
reefs are land
abo
are easily butubutu
reached by (kint
paddling
common (boundaries).
fishing methods curre
goon Most
include
angling Vonavona
with hook
(vaqara), divingmains within
(suvu), t
spearing
shells (hata). lamana
On average(open
most se
a
marine resourceopen-sea-facing
and consume fres
i
three times per week.
(mainland). With
Our long-term ethnographi
logical categories
15 years) describes
holapana howorVon
san
tualize their grass),
environmentkopi (lag
th
(Fig. 2). Pepeso translates
ovuku (river lite
mou
but is typically employed
(reef drop). as
Thes
that demarcates land-sea
morphology, terr
abi
interior mountain-ridge-tops
plant and o
animal
As is commonly found
different elsewhe
substra
land and sea ecological
ime (algaezones
spec
ceptualized as ngongoto (prim
ontologically dis
as aspects of ovalis),
an integrated
nelaka (w

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 Hum Ecol (2016) 44:33-46 37

Pepeso
l 1
Tutupeka Poana Toba Vuragarena
nainland) (lagoon) (barrier (outer barrier/
island) open sea

Sagauru
(reef)

Kulikuliana
(seagrass meadow)

Fig. 2 Cross-section schematic of a generalized pepeso in Vonavona, showing local environmental classifications and their approximate English
equivalents

patu pede (generic term for Turbinaria, Pavona, and of the tsunami described how the lagoon water first reced
Acropora corals), patu vinu (.Acropora corals), patu voa ed from the shoreline then returned as several one-meter

(.Pontes coral formations), tatalo (algae) and zalekoro (grav
el). Usually the dominant benthic types identified by local
people include onone, nelaka, kulikuliana, and patu. Impor
tantly, these dominant benthic types correspond very closely
to common marine science benthic categories of sand (onone),
silt (nelaka), coral (patu), and seagrass (kulikuliana).
The impact of the earthquake and tsunami on Vonavona
high pulses or waves. The waves swept into the lagoon,
flooding many areas and causing violent currents and ma
jor movements of rubbish- and silt-laden water. A long,
narrow island called Rokama, near the entrance of the
lagoon, was permanently cut in half by the tsunami.
Data Collection

Lagoon was significant. Although no villagers died in

Vonavona (nearly 60 lives were taken in other regions) it In 2010 we conducted 58 surveys in Kinamara (w = 28),
was a life-changing event that has had lasting impact includ Kinda (n = 6), and Saika (n = 24) where information was
ing extensive coastal flooding, changes to the local ecology, elicited to show the extent of villagers' IEK and nine
and major property damage, not to mention much human suf socioeconomic variables that, according to previous re
fering and psychological distress. Luckily, the villagers' first search, may influence levels of knowledge: age, gender,
reaction to the earthquake was to flee the coast and seek high number of years of local residence, level of education,
ground, a response that was common across the Western Prov type of occupation, days per week marine food is con
ince and saved many lives (Lauer 2012). Fearing another tsu sumed, days per week spent fishing, monthly income,
nami, Vonavona villagers spent 2 to 3 months living in tem and average weekly expenditure on processed food
porary shelters on high ground near the center of Parara Island (Table 1) (Zent 2013). The number of surveys conducted
and very few people ventured into the lagoon to fish or collect in each community was based on population size and the
marine resources. total sample size represents approximately 50 % of house
Lagoon and coastal areas were heavily impacted. Be holds in the three villages. A relatively even number of
cause of its position relative to the fault, the lagoon sank men and women from different age groups was chosen to
0.51 m (Taylor et al. 2008). This left many low-lying represent all ages over 18 years of age.
areas permanently flooded. The coastal areas of Parara A subsample of these individuals was asked additional
Island suffered much saltwater intrusion killing large questions about their social and expert networks. To generate
swaths of mangrove. Of the three study villages, our subsample for the social network study, we employed a
Kinamara was the most severally impacted with aboutsnowball sampling method where we initially asked 17 re
30 % of the village area permanently flooded, forcingspondents about their social and expert networks. This in
the community to rebuild its dock and relocate a number volved asking informants to name the seven most important
of public buildings such as the community hall and a persons in their lives, starting with the most important, outside
clinic. Fortunately, no houses were destroyed because of their household (for more details about this sampling
they were all located on higher ground. Local descriptions strategy see Atran et al. 2002). We determined this to be their

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 38 Hum Ecol (2016) 44:33-46

Table 1 Descriptive statistics of all

N= 58 N= 32

Variables Description Min Max Mean Std. Min Max Mean Std.
Dev. Dev.

Dependent
Agreement Agreement between indigenous knowledge 0 5 2.5 1.3 0 5 2.4 1.5
of abiotic and biotic substrates and a 2010
marine science survey.
Explanatory
Age Age of participant in years 24 75 44.7 12.6 31 75 49.7 11.4

Years in village Years the participant has lived in the village 1 75 19.3 18.4 4 75 22.0 20.0

Times per week Fishing Average number of fishing trips per week 0 6 2.4 1.9 0 6 2.5 2.0

Days per Week Fish Consumed Average number days a week a participant 1 7 2.7 1.9 1 7 2.8 2.1
eats fresh fish
Monthly Income (adjusted) Average monthly cash income 0 13000 1106.7 1854.3 0 13000 1275 2375

Money Spent on Processed Food Average amount of money spent on processed15 800 128.8 116.1 20 800 139.0 140.4

food per week

DegreeSN A participant's degree measurement
- - - -
1 11 4.9 2.4
for the social network
BetweennessSN A participant's betweenness measurement
- - - -
0 111.7 27.4 32.7
for their social network
Indegree_EN A participant's degree measurement
- - - -
0 10 1.7 3.0
for the expert network
BetweennessEN A participant's betweenness measurement
- - - -
0 45.7 4.8 11.7
for the expert network
N % of total N % of total

Occupation 1 = Farmer 28 48 % 14 44%

2 = Fisher 6 10 % 3 9 %

3 = Pastor 6 10 % 4 13 %

4 = Housewife 7 12 % 2 6 %

5 = Salaried work 6 10 % 6 19 %

6 = Other 5 9 % 3 9 %

Gender 1 = Male 32 55 % 21 66 %

2 = Female 26 45 % 11 34 %

Education 1 =None 0 0 % 0 0 %

2 = Up to 6th grade 44 76 % 23 72 %

3 = 6th-9th grade 7 12 % 5 16 %

4 = 9th- 11th grade 1 2 % 0 0 %

5 = High school graduate 0 0 % 0 0 %

6 = Trade school 4 7 % 2 6 %

7 = University 2 3 % 2 6 %

social network. We also asked people to indicate whom they To evaluate local knowledge about ecological change as
would turn to find out something they did not understand sociated with the 2007 tsunami, the survey included an elici
about the marine environment. We determined this to be their tation technique where a poster-sized color air photograph of
expert network. Due to time constraints, we were able to find the Repi study area was used as a visual tool to assist all 58
and interview only 15 of these individuals to represent the villagers in identifying benthic substrates. A 61 x 122 cm
total social network sample of 32 individuals. Since half of hard-copy map was created by digitally scanning and rectify
the population in a typical Solomon Islands village is under ing a color aerial photograph (1:25,000) of the southern
the age of 18 (Solomon Islands Government 2011), our social Vonavona lagoon on September 2, 1991. Five polygons were
network sample of 32 individuals was approximately 13 % drawn on the image to demarcate different sites within the
(32 out of 250) of villagers 18 years or older and one-third larger study area (Fig 3). Using the poster-sized map, we
(29 %) of village households (assuming each interviewee rep helped the informants orient themselves to the aerial perspec
resents one household). tive by encouraging them to recognize, identify and name

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 Hum Ecol (2016) 44:33-46 39

Fig. 3 Air photograph of r epi reef

Points are the site locations for mar

villages, islands and

documented the entire study site's other
abiotic and biotic substrates phys
map. The informants
before and after the 2007 tsunami. For eachwere
survey, we select the
and name the dominant
ed underwater benth
sample site locations by creating a grid using
language based on
GIS software their
that generated recolle
points every 60 m, for a total of
the benthos. Ideally
982 sampling points across the 253 hawe would
study area. The geo
the lagoon, but because
graphic coordinates high-r
of each sample point location were loaded
able we used into a Trimble Geoexplorer
the 1991 XT GPS receiver.
air Using the GPSphoto
interviews itreceiver,
becamea team consisting of a researcher apparent
and two trained,
mnemonic device or visual aid that enabled the informant to local research divers from nearby Roviana lagoon navigated
orient themselves in the seascape rather than something that by boat to each predetermined field point location and docu
was read or interpreted. In other words, informants did not mented the underwater habitats. At each site al-mxl-m PVC
simply interpret the various shades of color on the image, frame was lowered onto the seabed and the indigenously de
but instead the image served to jog the informants' memory fined substrate and dominant benthic habitat categories (the
about the study reef, an area that they cross frequently (on same categories used in the household survey mapping exer
average once or twice a week) during their fishing trips. cise) were recorded in the local language. If the sampled area
Through this elicitation technique ten dominant substrate had a mix of different habitats, the primary (dominant), sec
types were identified: ime (algae species), kuli gele (Enhalus ondary, and/or tertiary habitats were recorded.
acoroides), kuli ngongoto (primarily Thalassia hemprichii To compare the MS surveys with the responses provided
and Halophila ovalis), nelaka (silt), omomo (algae species), by local informants on the household survey, we used Arc/
onone (sand), patu mateana (dead coral) patu vinu (Acropora GIS 10 to run spatial queries that selected just those points
corals), patu voa (.Pontes coral formations) and tatalo (algae). inside the five study site areas assessed by the informants
For statistical analysis, we reduced these ten categories to four during the household survey. The five study sites contained
useful categories by combining the seagrass and algae species between three and 26 MS survey sample sites for a total of 59
into one category and the coral species into one category, samples (N= 59) each year. Using these data sets we measured
leaving us with the categories: coral, sand, seagrasses and the correspondence between local assessments of benthic
algae, and silt. types elicited through the image interpretation techniques
The three underwater MS surveys conducted by the same and our 2010 MS survey results. We analyzed only the 2010
research team in January 2006, July 2008 and May 2010 MS data set because that was the year in which individual

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 40 Hum licol (2016) 44:33—46

Table 2 Dominant abiotic and biotic benthic substrates recorded

villagers were surveyed. Al
during marine science visual surveys (N= 59 for each survey year)
sometimes given in both t
survey indicating
Survey Year a mixture
only the first response, rep
Study site 2006 2008 2010
type. Selecting only the dom
1 sand sand silt
is a common strategy in h
2 sand seagrass/algae sand
2013) because it provides
3 silt silt silt
any changes that occurred to
4 seagrass/algae seagrass/algae seagrass/algae
increasing the complexity o
5 sand sand sand
compared these two data set
if there was agreement betw
and the MS survey, then it
agreement Tukey post-hoc was
it test. All statistical analyses were run in R
coded as 0
five sites version 3.1.1 (2014).
producing our d
which rangedThe subsample
from (N= 32) of interview respondents who were 0 to 5, w
between indigenous
asked to provide information on social and expert networks knowle
allowed us to calculate two degree and two betweenness cen
trality measures using the social network analysis software
Data Analysis
package UCINet 6 (Borgatti et al. 2002). Degree centrality
identifies the direct connection between two individuals and
Three separate analyses
betweenness centrality were
offers a measure of the importance of
collected from household
individuals that lie in the shortest path between two individ su
was performed
uals. Social and expert with the
network information was used to con s
(iV= 58) to determine
struct two binary adjacency matrices (32 x 32) containingthe
only
agreement the names of those who were both interviewed
between IEKand named as and
point contacts. Those individuals who
agreement scorewere named but not as a
variable and nine socioeconomic variables were selected as interviewed were not included in this analysis. Data from the
independent variables.1 An ordinal logistic regression modelsocial network were treated symmetrically while the expert
was chosen to analyze the effects of the independent variables
network data were treated asymmetrically. The resulting mea
on a respondent's agreement score as this models the proba
sures served as four additional independent variables and were
bility that agreement is less than or equal to a given score, and
added to the variables used in the analysis of the larger sub
hence can be manipulated to provide the odds of a higher
sample and evaluated employing a method similar to that not
score. The equation for such a model is of the form ed above.

P(agree<j)
Iogit[P(agree<j)] = log
1 -P(agree<j)
Results
— Po + P\x\ + ••• + PfXt

The MS surveys showed that benthic substrates in two of the

where j-0, 1, 2, 3, 4 (the probability that agreement is<5 is
five surveys sites changed after the tsunami (Table 2). Site one
equal to one), and x represents an independent variable.
For the first step of the analysis, we conducted bivariate remained unchanged between years 2006 and 2008, but then
shifted from sand to silt in 2010. Site two changed from sand
screenings to consider each potential variable as the sole var
in 2006 to seagrass in 2008 and back to sand in 2010. Sites 3
iable in the model and a likelihood ratio test was run compar
ing each model to a null (intercept only) model. All indepen
5 showed no change over the study period. The agreement
between IEK assessments of substrate surfaces and the MS
dent variables found to be significant at the p< 0.1 level were
then placed in a multi-variable model to examine their signif surveys ranged between 0 (no agreement) to 5 (complete
icance when controlling for the other variables. These models
agreement) with a mean agreement of 2.5.

were optimized so that in order for a variable to remain in the Using a predetermined cutoff point of 90 % confidence

final model it had to be significant at a<0.05. As a final step,

level, a <0.1, for inclusion in a multi-variable analysis, the
a one-way analysis of variance test was run for which the bivariate screening for the full sample (jV=58) revealed that

independent variable occupation was the only significant var

' We considered using data transformation techniques on several vari
iable emerging from the screening analyses. This allowed for ables to provide more normal distributions, but the transformed data did
group-wise comparison of mean agreement scores using a not affect the outcome of our statistical analysis.

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 Hum Ecot (2016) 44:33-46 41

Table 3 Results of Bivariate

Screening of Independent 2010 (Af=58) 2010 (Af=32)
Variables tested for significance
Variable Name LR chi-square p-value LR chi-square p-value
in an ordinal logistic regression
model Age 2.052 0.152 1.031 0.310

Gender 0.189 0.664 0.83 0.362

Years in Village 1.829 0.176 2.36 0.124

Education 2.223 0.695 7.213 0.065"

Occupation 19.495 0.002* 20.593 0.001**

Times per week Fishing 0.588 0.443 0.573 0.449

Days per Week Fish Consumed 0.039 0.843 0.059 0.808

Monthly Income 0.015 0.902 0.052 0.819

Money Spent on Processed Food 1.531 0.216 3.547 0.060**

Degree (Social Network)

- -

0.624 0.430

Betweenness (Social Network) - -

0.634 0.426

Indegree (Expert Network)

- -
0.056 0.813

Betweenness (Expert Network)

- -
0.037 0.847

* For the 58 sample screening runs only Occupation proved to be significant

**For the 32 sample screening runs, education, occupation, and money spent on processed food were all found to
be significant at 90 % significance and so were included in a multi-variable model

only the variable occupation had a statistically significantused

re the occupational group 'fisher' as the reference
group to test if fishers had the highest IEK-MS agreement
lationship to the dependent variable IEK-MS agreement score.
score due to the large amount of time they spend
This relationship was found to be a strong one, \2 (5)= 19.49,
p- 0.002. Because only that independent variable showedinteracting
a with the marine environment. Our results,
however, indicated that the odds of salaried workers hav
significant relationship to the dependent variable, no multi
variable work was done with the larger sample. ing a higher IEK-MS score were more than 28 times those
of individuals who fish for a living (Table 5). Importantly,
The bivariate screening for the reduced sample (N- 32)
found that three socioeconomic variables, 'education,'salaried workers also fished quite regularly even though
'occupation,' and 'money spent on processed food' show
they had salaried employment. On average they fished of
2.5 times per week (N= 58), which was just below the
a significant relationship to IEK-MS agreement at the
90 % confidence level (Table 3). These variables were
mean (2.6 times per week) of all respondents (Table 6).
placed in a multi-variable model for which a backward We further analyzed the differences in mean scores for
step-wise reduction was performed. As with the larger
the occupation groups in the larger sample (N= 58) using
sample, a final model showed 'occupation' to be highly
a one-way ANOVA test (Table 5). Mean agreement scores
significantly related to IEK-MS agreement, were
\2 found to be significantly related to occupation, F (5,
(5) = 20.593,/? = 0.001 (Table 4). 52) = 3.95, p = 0.004. A post-hoc Tukey examination of
pair-wise differences found that the groups for which the
Further examination of the occupation variable indicat
mean differed significantly at p< 0.5 were Salaried/Other,
ed that of the six different self-reported occupations un
Salaried/Farmer, and Salaried/Pastor, and Salaried/
dertaken by informants (fisher, farmer, pastor, housewife,
Housewife at/? <0.1. There was no statistically significant
salaried worker, and/or other), in individual comparisons,
difference in means between salaried workers and fisher
only villagers with salaried employment had significantly
men, which was likely due to an outlier (see Fig. 4 and
higher odds of selecting benthic habitats that agreed with
the MS survey than the reference group of fisherman. Wethe Discussion section). Moreover, the higher p-value may

Table 4 Reduction of multi

variable model with three Run 1: Run 2: Run 3:

variables (N= 32)

Variable Chi-Square Pr>ChiSq Chi-Square Pr>ChiSq Chi-Square Pr>ChiSq
Education 5.44 0.1421 5.61 0.132 20.59 0.001

Occupation 16.05 0.0067 18.99 0.0019

SpendProcessed 0.28 0.5946

Food

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 42 Hum Ecol (2016) 44:33^16

Table 5 Odds ratios of obtaining

Occupation Odds ratio p-val Adj. p-val Mean Agreement Std Dev
a higher agreement score when
compared to the reference group,
Farmer (n = 28) and0.786 0.77 0.98 2.36 1.19
fishermen, mean agreement
scores for
Fisherman («each
= 6) occupation
reference group
2.67 1.37 - -

(JV=58) 0.318 0.29 0.98 1.83 1.47

Pastor (n = 6)
Housewife (n = 5) 1.023 0.98 0.98 2.57 0.98

Salaried worker (n = 6) 28.526 0.01 0.05 4.33 0.82

Other (n = 5) 0.258 0.21 0.98 1.60 1.14

* Hochberg adjustment is made to the p-values to account for multiple comparisons

be a function of the small sample size and in such cases it (Bicker etal. 2004; Reyes-Garcia etal. 2005). Factors associ
is often better to look at the various means and standard ated with modernity such as years of school, academic skills,
deviations (Table 7). fluency in a non-local language, household income, and dis
It is important to note that the bivariate screening tance
for to markets or towns, have all been shown to be nega
the reduced sample (N= 32) found that none of the social
tively associated with local ecological knowledge (Reyes
network variables had a statistically significant relation
Garcia et al. 2005). The erosion of knowledge has been ex
ship with the dependent variable. We acknowledge, plained
how based on the idea that acculturation and modernization
ever, that time constraints prevented the collection ofdisplace
so ecological knowledge as people acquire the skills
cial network data from every individual and, therefore, a
needed to live in a more market-centered economy.
more thorough analysis of the global pattern of connec However, recent analyses suggest that the relationship is
tions between nodes. Of the two measures of centrality
more complex, and that IEK responds in different ways to
that were calculated, we have much more confidence inthe forces of globalization (Gomez-Baggethun and Reyes
the degree measurement than the betweenness measure Garcia 2013). In some contexts, wage labor may not displace
ment because degree is not affected by an incomplete local ecological knowledge. Such a finding was reported in a
network sample as it measures direct interaction between study among the Tsimane' of the Bolivian Amazon who were
two individuals. Our betweeness measurement, on the employed in wage labor involving the harvest and sale of
other hand, has limited explanatory power when calculat forest products but retained their ethnobotanical knowledge
ed for an incomplete network and should be viewed as (Reyes-Garcia et al. 2007). Likewise, in Roviana Lagoon,
only a rough guideline. modern knowledge about non-subsistence cash crops was in
tegrated into communities without eroding traditional

Discussion

Two particular findings from this research prompt discussion.

First, our results suggest that Solomon Islanders who earn
salaries tend to be the most adept at detecting ecological
changes associated with infrequent, large-scale disturbances
like tsunamis. This is contrary to research linking market in
tegration and increased levels of modernity to the loss of IEK

Table 6 Days per week

fishing by occupation Occupation Mean days per
(N= 58) week fishing

Farmer 2.1 (m = 28)

Fisherman 4.2 (n = 6)
Pastor 1.7 (n = 6)
T I | | | p
Housewife 1.9 (n = 7) 1 2 3 4 5 6
Salaried worker 2.5 (« = 6) Occupation
Other 3.1 (n = 5) Fig. 4 Boxplot s
Average days fishing 2.6 (N= = 58) different occu
for all participants Pastor; 4 - Hou
outlier in Occup

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 Hum Ecol (2016) 44:33^6 43

Table 7 Tukey post-hoc

salaried workers, much more compariso
than most villagers, to glob
scores between Occupation groups
al knowledge about current affairs and government poli
Occupation Difference lower CL Upper CL p adj cies and initiatives.
Comparison Interestingly, results of the statistical analysis revealed an
outlier with a considerably higher IEK-MS agreement score
6-5 -2.733 -4.854 -0.613 0.005'
than others in her occupation group. While it is not unusual to
5-1 1.976 0.401 3.551 0.006'
remove or manipulate outliers for statistical analysis, one of
5-3 2.500 0.478 4.522 0.007'
the main points of this study was to understand variability in
5-4 1.762 -0.186 3.710 0.098'
knowledge, thus we believe that removal of data points simply
5-2 1.667 -0.355 3.688 0.162
because they are extreme in comparison to others does not
6-2 -1.067 -3.187 1.054 0.673
necessarily mean that the value is invalid or erroneous. While
6-4 -0.971 -3.022 1.079 0.726
outliers may change the nature of the statistical relationship
6-1 -0.757 -2.457 0.943 0.774
between two variables, particularly in a small sample size or if
3-2 -0.833 -2.855 1.188 0.825
they are illegitimate (i.e., due to sampling error), we decided to
4-3 0.738 -1.210 2.686 0.870
retain it to demonstrate the importance of understanding the
3-1 -0.524 -2.099 1.051 0.921
social context of each of these data points. In this particular
1-2 -0.310 -1.885 1.266 0.992
case, the outlier is a middle-aged woman who self-reported
4-1 0.214 -1.265 1.694 0.998
that fishing is her primary occupation. Importantly, she was
6-3 -0.233 -2.354 1.887 0.999
the only respondent in this occupational category born in
4-2 -0.095 -2.043 1.853 1.000
Munda, one of the largest towns in the region, and, as
discussed above, a source of other "types" of knowledge.
1 - Farmer; 2 - Fisherman; 3 - Pastor; 4 - Housewife; 5 - Salaried
worker; 6 - Other Her link to a more urbanized center may have exposed her
*p<0.1 to global knowledge about the effects of the tsunami, suggest

**p< 0.05 ing that she may not be a statistical anomaly, but rather an

***/><0.01 important case for understanding IEK in the region.

One kind of global knowledge that salaried workers (and
possibly the woman mentioned above who was bom in Mun
ethnobotanical knowledge (Furusawa 2009). In these cases, da) would have been exposed to after the 2007 tsunami was
the introduction of new knowledge did not necessarily dis the national news media. Numerous radio and newspaper re
place local ecological knowledge; rather, if people had ports emphasized that the tsunami had a significant impact on
sustained contact with the natural environment, they main the reefs and marine habitats of the western Solomon Islands.
tained their IEK. Importantly, Vonavona villagers who earn These reports generally did not give details about changes in
salaries maintain a regular relationship with the marine envi specific areas, but they could have made salaried individuals,
ronment. On average individuals earning salaries fished who would more frequently be exposed to these news reports,
2.5 days per week, which is near the mean for the entire sam more aware that tsunamis cause ecological change. We should
ple (2.6). point out that, in the western Solomons, IEK about tsunamis is
Although individuals who earn salaries fish frequently, minimal. Tsunami-related knowledge is not encoded in the
this does not explain why they would be more adept at language (i.e., the Roviana language does not have a word
detecting change than other villagers who also fish regu for 'tsunami') and there is no known oral history or myths
larly. This may be due to the unique position salaried recounting large destructive waves occurring after earth
workers play in Solomon Island communities. Most sala quakes (Lauer 2012). The lack of local knowledge is attribut
ried work requires that people leave their home villages able to the rarity of tsunamis in this region. Although earth
and live in urban areas like Gizo and Honiara. In our quakes are frequent in the western Pacific, there has not been a
sample, however, individuals earning salaries largelived inor greater magnitude) earthquake, capable of gener
(7.0
their villages. They included two schoolteachers, ating
two gova tsunami, in over 100 years (Taylor et al. 2008). It
ernment workers, one mechanic, and one person standswho
to reason, then, that exposure to global knowledge
worked at a small, nearby eco-resort. Even though about the tsunami (especially through print and television
these
people permanently resided in the villages, their news) and its effects could have sensitized salaried workers
jobs re
quired them to take frequent trips to town centersto ecological
like changes associated with the tsunami more than
Munda or Gizo to collect pay checks. In addition, other
teachers
villagers who rely solely on IEK.
The second
and government employees occasionally travel to town to important finding is that measures of net
work centrality
receive training, and mechanics need access to markets to were not positively associated with eco
logical
buy spare parts. These visits to urban centers exposechange detection abilities. This result contrasts

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 44 Hum Ecol (2016) 44:33 46

with a large the village to contacts in of

body town. But salaried
socialworkers did not ne
structural mention anyone outside of theinfluences
position three villages during the net
edge disperses through
work survey interviews. a com
Reyes-GarciaIf weet conceptualizeal. 2013).
habitat change detection as a situated Im
studies conceive of
practice, it is reasonable to suggestknowled
that a person's aptitude
information that flow
might not be influenced by their position in a throug
social or expert
et al. 2009), anetwork (see Boster 1986:434). Changes in benthic habitats
conceptualizatio
mise that would unlikely be a broad topic of conversation
knowledge is because a the
corp
instructionsway inthat
which Vonavona villagersare distin
generate knowledge about
day-to-day activities. This
their marine environment is grounded in the actual activity of c
expressed quite clearly
fishing. To be a successful fisher in the Solomon Islands, in a
indigenous habitat
knowledge,
identification and monitoring is crucial. Each habitat w
"cumulative has different species associated
body of with it that follow certain en
knowled
through vironmental parameters such as changes
generations by in tide andcultu
moon
relationshipphases, of living
and each requires being
different baits and gear types (see
one another and with their environment" (Berkes Lauer and Aswani 2009 for a detailed description). Habitat
1993:3). A growing body of literature, however, questions change is common in these dynamic environments, and fish
this conceptualization of knowledge. Scholars from a ers must develop the ability to perceive habitat changes and
number of disciplines disagree with the assumption that adapt their practices accordingly (Lauer and Aswani 2010).
knowledge is "in the head," separated from the lived-in Even though fishers most certainly memorize certain aspects
world (Ellen et al. 2000; Ingold 2000; Kightley et al. of the local environment like the name and location of fishing
2013; Lave 1988; Richards 1993; Scott 1998; Stone grounds, and other nomenclature (e.g., fish names and behav
2007; Zarger 2011). In fact, Solomon Islanders' own con iors, gear types), this information is continually regenerated
ceptions of knowledge parallel this view in that they privwithin the context of frequent fishing forays and inseparable
from the process of perceiving subtle variations in environ
ilege the perceived outcomes of practical activities rather
than the metaphysical or abstract (Hviding 1996; Lauer
mental conditions as they unfold while out on the lagoon
and Aswani 2009). A variety of labels such as tacit fishing. Fishing is almost never taught formally or transmitted
knowledge, metis, situated practices, or skill have through oral instruction; rather it is acquired by youngsters
emerged that stress how knowledge and knowing is intrin through watching and fishing with others who are more adept.
sically dynamic, performed, and improvised, and cannot By the time a fisher is in his mid- to late-teens, he or she has
be separated from the context in which they are applied. developed the sensibilities and dispositions necessary to detect
From this processural or practice perspective, knowledge currents, wind conditions, bait fish movements, or habitat
is never fully stable and durable. It is continually regen change, but this knowledge is not a topic of conversation
erated, a process that entails what has been described as and is not necessarily orally transmitted. Moreover, as stated
"situated messiness" (Tumbull 2000:39), where new in above, tsunami-induced changes are typically not within the
formation about the environment or improvements in horizon of a Solomon Island fisher's environmental sensibil
technique intermingle with more established modalities ity. It was the exposure of salaried workers to the global dis
of knowing. course about the effects of tsunamis that tuned them into the

This alternative conception of knowledge encourages us to possibility that environments might change after tsunamis. A
approach the ecological change detection we assessed in situated-practice approach to knowledge provides a theoreti
Vonavona Lagoon not as bits of information about the envi cal basis to understand not only why a fisher's social network
ronment, but rather a situated practice. Therefore, villagers (in this article defined as important social acquaintances) has
who earn salaries and, as a result, were exposed to media little influence on his or her skills, but also how IEK and
reports about ecological change associated with the tsunami global knowledge can intermingle.
were not receivers of specific information about the effects of When we consider the results of our social network analy
the tsunami on the local marine ecology, rather their reper sis from a practice-based approach to IEK, it suggests that a
toires of understanding were expanded, altering their percep different kind of social network may be more important, such
tion of the environment in a way that sensitized them to the as a fisher's fishing partners. It is within what Lave and
possibility of tsunami-induced ecological changes. Our results Wenger (1991) call 'communities of practice' that skills and
suggest that this process of sensitization was driven by sala expertise are maintained and regenerated. For fishers, their
ried workers' exposure to print, TV, and online news media primary community of practice would be the group of fishers
rather than through their social networks. It is quite possible
with whom they fish on a regular basis. An important avenue
that salaried workers' social networks might extend outside of future research would be to explore the effect these various

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 Hum Ecol (2016) 44:33-46 45

kinds of social (2011). Linking Indigenous and Scientific Knowledge

networks of Climate
have o
Change. Bioscience 61: 477-484.
and modalities of knowing.
Aswani, S. (1998). Patterns of Marine Harvest Effort in Southwestern
To conclude, we highlight
New Georgia, Solomon Islands: Resource Management or an i
Our results suggest that some
Optimal Foraging? Ocean and Coastal Management 40: 207-235.
at detecting change and
Aswani, S., and Lauer, M. (2014). Indigenous there
people's Detection of
useful in the Rapid Ecological Change. Conservationof
design Biology 28: 820-828.
resou
Aswani, S., M. Lauer, P. Weiant, L. Geelen, and S. Herman. (2004). The
disaster risk-reduction mit
Roviana and Vonavona marine resource management project, final
against importing assumptions
report, 2000-2004. Department of Anthropology, University of
knowledge and about
California, Santa Barbara. who are
experts. The Atran, S., and Medin,
fact D. L. (2008). The Native
that Mind and the Cultural
salaried
are Construction
exclusively of Nature. Life and Mind. MIT Press, Cambridge.
fishers were
Atran, S., Medin, D., Ross, N., Lynch, E., Vapnarsky, V., Ek, E. U., Coley,
ecological change
after the t
J., Timura, C., and Baran, M. (2002). Folkecology, Cultural
spread assumptions that local
Epidemiology, and the Spirit of the Commons: A Garden
most experience
Experiment engaging
in the Maya Lowlands, 1991-2001. Current wit
and Wagner Anthropology 43: 421-450.Even
2003). local
Berkes, F. 1993. Traditional Ecological Knowledge in Perspective. In
consider salaried workers as f
Inglis, J. (ed.), Traditional Ecological Knowledge Concepts and
individuals would have been m
Cases. International Program on Traditional Ecological
on our 'expert network'
Knowledge: International Development Research Centre, socia
Ottawa,
volved in Ont., Canada, pp. 1-10.
resource management
cognizant Berkes, F., Colding,
the J., and Folke, C. (2000). Rediscovery
these of kinds of Traditional of
Ecological Knowledge as Adaptive Management. Ecological
edge generation in indigenous
Applications 10: 1251-1262.
edge may put Berlin,
be put
B. (1992). Ethnobiological into pract
Classification: Principles of
intersection of local
Categorization of Plants and Animalsand glob
in Traditional Societies.
IEK as a complex,
Princeton University multifacet
Press, Princeton.

more Bicker, A.,sensitive,

inclusive, Sillitoe, P., and Pottier, J. (2004). Investigating Local
and,
Knowledge: New Directions, new Approaches. Ashgate, Aldershot.
agement that policies contribut
Bodin, O., and Prell, C. (2011). Social Networks and Natural Resource
communities over the
Management: Uncovering the Social Fabric long-ter
of Environmental
Governance. Cambrdige University Press, Cambridge.
Borgatti, S. P., M. G. Everett, and L. C. Freeman. (2002). Ucinet for
Acknowledgments Windows: Software Many
for social network analysisthanks
for supportingBorgatti,
this S. P., Mehra, A., research.
Brass, D. J., and Labianca, G. (2009). NetworkIn p
Tungama Gua, and Analysis in theLawrence
Social Sciences. Science 323: 892-895. Buka
Campanella and Ben Nugent
Boster, J. S. (1986). for
Exchange of Varieties and Information thei
Between
Campbell for
her Aguaruna
assistance with
Manioc Cultivators. American Anthropologist 88: 428 sta
Sore, Permanent 436.Secretary of the S
ment, Conservation,
Brokensha, D., Warren, D. and Meteoro
M., and Werner, O. (1980). Indigenous
input of three anonymous reviewe
Knowledge Systems and Development. University Press of
Human Dimensions
America, Washington. and Social
#0827022) and San Diego State Univ
Davis, A., and Wagner, J. R. (2003). Who Knows? On the Importance of
support for this research.
Identifying "Experts" When Researching Local Ecological
Knowledge. Human Ecology 31: 463-489.
Compliance With Ethical
Drew, J. (2005). Standar
Use of Traditional Ecological Knowledge in Marine
Conservation. Conservation Biology 19: 1286-1293.
Conflict of interest The authors h
Ellen, R. F. (ed.) (2007). Modem Crises and Traditional Strategies: Local
proval for this research was granted
Ecological Knowledge in Island Southeast Asia. Berghahn Books,
dation. InformedNew consent
York. was obtain
this study.
Ellen, R., Parkes, P., and Bicker, A. (eds.) (2000). Indigenous
Environmental Knowledge and its Transformations: Critical
Anthropological Perspectives. Harwood Academic Publishers,
Amsterdam.

Escobar, A. (1999). After Nature: Steps to an Anti-Essentialist Political

References
Agrawal, A. (1995). Dismantling the Divide Between Indigenous and
Scientific Knowledge. Development and Change 26: 413-439.
Alexander, C., Bynum, N., Johnson, E., King, U., Mustonen, T., Neofotis,
P., Oettle, N., Rosenzweig, C., Sakakibara, C., and Shadrin, V.
Ecology. Current Anthropology 40: 1-30.
Furusawa, T. (2009). Changing Ethnobotanical Knowledge of the
Roviana People, Solomon Islands: Quantitative Approaches to its
Correlation With Modernization. Human Ecology 37: 147-159.
Gomez-Baggethun, E., and Reyes-Garcia, V. (2013). Reinterpreting
Change in Traditional Ecological Knowledge. Human Ecology 41:
643-647.

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms
 46 Hum Ecol (2016) 44:33-46

Goodman, A., A Deadlock? Quantitative Research

Purkis, J. S. fromJ.,
a Native Amazonian
and Phi
Sensing. Guide for
Society. Human A Mapping,
Ecology 35: 371-377. Mo
Springer, New Reyes-Garcia, V., Molina, J. L., Calvet-Mir, L., Aceituno-Mata, L.,
York.
Green, A., P. Lokani, W.
Lastra, J. J., Ontillera, R., Parada,Atu, P.
M., Pardo-de-Santayana, M., Ram
(2006). Solomon Rigat, Islands
M., Valles, J., and Gamatje, T.marine
(2013). "Tertius Gaudens": asse
survey conducted GermplasmMay Exchange Networks and 13 to
Agroecological June
Knowledge
Country Report No.Among Home1/06. Gardeners in the Iberian Peninsula. Journal of
Gupta, A. (1998). Ethnobiology and Ethnomedicine 9: 53.
Postcolonial Developm
of Modern India.
Richards,Duke University
P. (1993). Cultivation: Knowledge or Performance? In Hobart, Pr
Hays, T. E. (1976). M. An Empirical
(ed.), An Anthropological Critique ofDevelopment: The GrowthMetho
Categories in Ethnobiology.
of Ignorance. Routledge, London, pp. 61-78. America
Hobart, M. (ed.) Ross,
(1993). An
N. (2002). Lacandon-Maya Intergenerational Anthropo
Change and the
The Growth of Ignorance.
Erosion of Folkbiological Knowledge. In Stepp,Routledge
J. R., Wyndham,
Hviding, E. (1996). E. S., andGuardians of
Zarger, R. K. (eds.), Ethnobiology and Biocultura! Mar
Politics in Maritime
Diversity. International SocietyMelanesia.
of Ethnobiology, Athens, pp. 585 U
Honolulu. 592.
Ingold, T. (2000). The Perception of the Environment: Essays onSchultes, R. E. (1994). Burning the Library of Amazonia. The Sciences
Livelihood, Dwelling and Skill. Routledge, London. 34:24-31.
Katona, Z., Zubcsek, P. P., and Sarvary, M. (2011). Network Effects and Scott, J. C. (1998). Seeing Like a State: How Certain Schemes to Improve
Personal Influences: The Diffusion of an Online Social Network. the Human Condition Have Failed. Yale University Press, New
Journal of Marketing Research 48: 425-443. Haven.
Kightley, E., Reyes-Garcia, V., Demps, K., Magtanong, R., Ramenzoni,
Solomon Islands Government. (2011). Report on the 2009 population
V., Thampy, G., Gueze, M., and Stepp, J. (2013). An Empirical and housing census: Basic tables and census description.
Comparison of Knowledge and Skill in the Context of Traditional Solomon Islands National Statistics Office
Ecological Knowledge. Journal of Ethnobiology and Spoon, J. (2011). The Heterogeneity of Khumbu Sherpa Ecological
Ethnomedicine 9: 71. Knowledge and Understanding in Sagarmatha (Mount Everest)
Lauer, M. (2012). Oral Traditions or Situated Practices? Understanding National Park and Buffer Zone, Nepal. Human Ecology 39: 657
how Indigenous Communities Respond to Environmental Disasters. 672.
Human Organization 71: 176-187. Stone, G. D. (2007). Agricultural Deskilling and the Spread of
Lauer, M., and Aswani, S. (2009). Indigenous Ecological Knowledge as Genetically Modified Cotton in Warangal. Current Anthropology
Situated Practices: Understanding fishers' Knowledge in the 48: 67-103.
Western Solomon Islands. American Anthropologist 111:317-329. Taylor, F. W., Briggs, R. W., Frohlich, C., Brown, A., Hornbach, M.,
Lauer, M., and Aswani, S. (2010). Indigenous Knowledge and Long Papabatu, A. K., Meltzner, A. J., and Billy, D. (2008). Rupture
Term Ecological Change: Detection, Interpretation, and Responses Across arc Segment and Plate Boundaries in the 1 April 2007
to Changing Ecological Conditions in Pacific Island Communities. Solomons Earthquake. Nature Geoscience 1: 253-257.
Environmental Management 45: 985-997. Tumbull, D. (2000). Masons, Tricksters and Cartographers: Comparative
Lave, J. (1988). Cognition in Practice: Mind, Mathematics, and Culture in Studies in the Sociology of Scientific and Indigenous Knowledge.
Everyday Life. Cambridge University Press, Cambridge. Harwood Academic, Amsterdam.
Lave, J., and Wenger, E. (1991). Situated Learning: Legitimate Peripheral
Vandebroek, I., and Balick, M. J. (2012). Globalization and Loss of Plant
Participation Learning in Doing. Cambridge University Press, Knowledge: Challenging the Paradigm. PLoS One 7, e37643.
Cambridge. Wasserman, S., and Faust, K. (1994). Social Network Analysis: Methods
Mercer, J., Dominey-Howes, D., Kelman, I., and Lloyd, K. (2007). The and Applications. Cambridge University Press, Cambridge.
Potential for Combining Indigenous and Western Knowledge in Zarger, R. (2011). Learning Ethnobiology. In Anderson, E. N., Pearsall,
Reducing Vulnerability to Environmental Hazards in Small Island D., Hunn, E., and Turner, N. (eds.), Ethnobiology. Wiley-Blackwell,
Developing States. Environmental Hazards 7: 245-256. Hoboken, pp. 371-387.
Ohmagari, K., and Berkes, F. (1997). Transmission of Indigenous Zarger, R., and Stepp, J. (2004). Persistence of Botanical Knowledge
Knowledge and Bush Skills Among the Western James Bay Cree Among Tzeltal Maya Children. Current Anthropology 45:413-418.
Women of Subarctic Canada. Human Ecology 25: 197-222. Zent, S. (2009). A Geneology of Scientific Representations of Indigenous
Pickering, A. (1995). The Mangle of Practice: Time, Agency, and Knowledge. In Heckler, S. (ed.), Landscape, Process, and Power:
Science. University of Chicago Press, Chicago. Re-Evaluating Traditional Environmental Knowledge. Berghahn
Reyes-Garcia, V., Vadez, V., Byron, E., Apaza, L., Leonard, W. R., Perez, Books, New York, pp. 19-67.
E., and Wiikie, D. (2005). Market Economy and the Loss of Folk Zent, S. (2013). The Processsural Perspectives on Traditional
Knowledge of Plant Uses: Estimates from the Tsimane' of the Environmental Knowledge: Continuity, Erosion, Transformation,
Bolivian Amazon. Current Anthropology 46: 651-656. Innovation. In Ellen, R., Lycett, S. J., and Johns, S. E. (eds.),
Reyes-Garcia, V., Vadez, V., Huanca, T., Leonard, W. R., and McDade, T. Understanding Cultural Transmission in Anthropology: A Critical
(2007). Economic Development and Local Ecological Knowledge: Synthesis. Berghahn Books, New York, pp. 213-265.

Springer

This content downloaded from

132.174.250.76 on Wed, 20 Apr 2022 06:27:33 UTC
All use subject to https://about.jstor.org/terms