15.2 Estimating the Date of First Settlement
253
settlement) gives 5,457, the factor by which the total pa area in 1445 has increased
from the area of the founding settlement and so the factor by which the total pa
population has increased from the founding population.
For an early first-wave settlement in 1500 BC close to the beginning of the Lapita
migration period (3,600–3,000 BP) and fully occupied pa dating to 1445, this factor corresponds to an average growth rate of 0.293% p.a. This calculation has the
advantage of being free of the uncertainties inherent in the use of Naroll’s method.
We note that without the double-counting correction, the corresponding result is
0.300% per annum. The smallness of the discrepancy is not surprising. Removing
all small pa only means removing 12.2% of the total area of all pa. And we note
that all the pa extracted as a double-counting correction are all well below X 0 which
is the mode of the double-tail distribution, so that the correction does not change
the mode of the distribution. The structure of the distribution is only partly modified
because the mode is intact and it is the mode that is relevant to the size of the founder
population.
For a late first settlement in 1000 BC at the end of the Lapita migration period,
the population growth rate to AD 1445 would be 0.353% per annum, a significantly
higher rate and so inherently less likely for tribal people who for most of their existence in New Zealand were settled hunter/gatherers. This suggests the likelihood
that the first settlement of New Zealand occurred closer to the beginning than the
end of the Lapita period.
For a very early second-wave settlement in 400 BC our estimate is for an average
population growth rate of 0.468% per annum. This high growth rate for a secondwave settlement immediately establishes it as implausible. It is improbable that such
a high growth rate could be sustained for 1,844 years for a population that for
nearly 1,500 years of that span led a settled hunter/gatherer existence. This result
makes it immediately clear that the first settlement of New Zealand was a first-wave
colonization. The calculations above, moreover, establish that the first settlement
of New Zealand was most likely an early Lapita-age colonization, one that took
place perhaps 200–400 years before the settlement of Western Polynesia and, given
recent late datings for first settlements in Eastern Polynesia, perhaps as much as
1,900 years before Eastern Polynesia was settled. This perception represents a major
change in the way in which the settlement sequence in the Pacific is viewed.
As noted in Chapter 14, a founding population of the order of 50 is often assumed
by prehistorians, sometimes with the connotation of accidental colonization. This
number has the disadvantage of being well below Flannery’s viability level of 500
for a mammalian population [6]. Suppose we were to assume a founding population
of only 50 in 1500 BC. Taken with the estimate of a pa population of 1,561,000 in
1445, this would give a completely unrealistic average population rate of increase of
1.00% p.a. as against the 0.29% p.a. for an initial population of 289 at the same date.
Sustained for 2,700 years for a settled hunter/gatherer population, a growth rate of
1% p.a. is untenable. This gives support to our estimate of the size of the founding population. Further, a first-wave colonization around 1500 BC with a founding
population of 289, presumably travelling in a fleet of canoes, would have been
almost certainly deliberate.
253
settlement) gives 5,457, the factor by which the total pa area in 1445 has increased
from the area of the founding settlement and so the factor by which the total pa
population has increased from the founding population.
For an early first-wave settlement in 1500 BC close to the beginning of the Lapita
migration period (3,600–3,000 BP) and fully occupied pa dating to 1445, this factor corresponds to an average growth rate of 0.293% p.a. This calculation has the
advantage of being free of the uncertainties inherent in the use of Naroll’s method.
We note that without the double-counting correction, the corresponding result is
0.300% per annum. The smallness of the discrepancy is not surprising. Removing
all small pa only means removing 12.2% of the total area of all pa. And we note
that all the pa extracted as a double-counting correction are all well below X 0 which
is the mode of the double-tail distribution, so that the correction does not change
the mode of the distribution. The structure of the distribution is only partly modified
because the mode is intact and it is the mode that is relevant to the size of the founder
population.
For a late first settlement in 1000 BC at the end of the Lapita migration period,
the population growth rate to AD 1445 would be 0.353% per annum, a significantly
higher rate and so inherently less likely for tribal people who for most of their existence in New Zealand were settled hunter/gatherers. This suggests the likelihood
that the first settlement of New Zealand occurred closer to the beginning than the
end of the Lapita period.
For a very early second-wave settlement in 400 BC our estimate is for an average
population growth rate of 0.468% per annum. This high growth rate for a secondwave settlement immediately establishes it as implausible. It is improbable that such
a high growth rate could be sustained for 1,844 years for a population that for
nearly 1,500 years of that span led a settled hunter/gatherer existence. This result
makes it immediately clear that the first settlement of New Zealand was a first-wave
colonization. The calculations above, moreover, establish that the first settlement
of New Zealand was most likely an early Lapita-age colonization, one that took
place perhaps 200–400 years before the settlement of Western Polynesia and, given
recent late datings for first settlements in Eastern Polynesia, perhaps as much as
1,900 years before Eastern Polynesia was settled. This perception represents a major
change in the way in which the settlement sequence in the Pacific is viewed.
As noted in Chapter 14, a founding population of the order of 50 is often assumed
by prehistorians, sometimes with the connotation of accidental colonization. This
number has the disadvantage of being well below Flannery’s viability level of 500
for a mammalian population [6]. Suppose we were to assume a founding population
of only 50 in 1500 BC. Taken with the estimate of a pa population of 1,561,000 in
1445, this would give a completely unrealistic average population rate of increase of
1.00% p.a. as against the 0.29% p.a. for an initial population of 289 at the same date.
Sustained for 2,700 years for a settled hunter/gatherer population, a growth rate of
1% p.a. is untenable. This gives support to our estimate of the size of the founding population. Further, a first-wave colonization around 1500 BC with a founding
population of 289, presumably travelling in a fleet of canoes, would have been
almost certainly deliberate.