DOI article:
Sullivan, M. E.; Hughes, Philip J.; Golson, J.: Prehistoric garden terraces in the eastern highlands of Papua New Guinea
DOI Page / Citation link:https://doi.org/10.11588/diglit.49002#0215
202
SULLIVAN, HUGHES, GOLSON
PAPUA NEW GUINEA
The present drainage system has dissected
these sediments down to solid Tertiary
bedrock, and the terraced slopes lie within
the old basin fill. The terrain in which the
terraces occur was mapped as the Abiera
Land System by Haantjens et al (1970). The
benches which make up one land unit of this
land system are the features identified as the
terraces in this study.
The landscape is largely covered by an-
thropogenic grasslands which are typically 1
m high and consist of a wide range of trailing
and sub-erect grasses (Robbins 1970). These
grasslands have replaced the former cover of
lower montane forest, remnants of which
occur on the crests and upper slopes of the
surrounding hills.
The Terraces
The results of preliminary geomorphological
and archaeological investigations of the ter-
races have been reported by Sullivan,
Hughes and Golson (1986) and Sullivan and
Hughes (1987). Much of the argument in
these papers was directed to showing that
previous descriptions which suggested a
natural origin for the features in question
could not be sustained. These previous de-
scriptions all sought to explain the strong,
horizontal element in the structures, their
superficially continuous and regularly
spaced layering and their limited areal oc-
currence within the valley landscape. The
explanations offered were: benchlike slumps
(Heyligers and McAlpine 1971), Pleistocene
lake beds (Dow and Plane 1965, repeated by
Tingey and Grainger 1976), and differential-
ly eroded layers of bedrock of varying hard-
ness (Rogerson et al 1982).
We do not wish to repeat the reasons why
SULLIVAN, HUGHES, GOLSON
PAPUA NEW GUINEA
The present drainage system has dissected
these sediments down to solid Tertiary
bedrock, and the terraced slopes lie within
the old basin fill. The terrain in which the
terraces occur was mapped as the Abiera
Land System by Haantjens et al (1970). The
benches which make up one land unit of this
land system are the features identified as the
terraces in this study.
The landscape is largely covered by an-
thropogenic grasslands which are typically 1
m high and consist of a wide range of trailing
and sub-erect grasses (Robbins 1970). These
grasslands have replaced the former cover of
lower montane forest, remnants of which
occur on the crests and upper slopes of the
surrounding hills.
The Terraces
The results of preliminary geomorphological
and archaeological investigations of the ter-
races have been reported by Sullivan,
Hughes and Golson (1986) and Sullivan and
Hughes (1987). Much of the argument in
these papers was directed to showing that
previous descriptions which suggested a
natural origin for the features in question
could not be sustained. These previous de-
scriptions all sought to explain the strong,
horizontal element in the structures, their
superficially continuous and regularly
spaced layering and their limited areal oc-
currence within the valley landscape. The
explanations offered were: benchlike slumps
(Heyligers and McAlpine 1971), Pleistocene
lake beds (Dow and Plane 1965, repeated by
Tingey and Grainger 1976), and differential-
ly eroded layers of bedrock of varying hard-
ness (Rogerson et al 1982).
We do not wish to repeat the reasons why