244
14 New Zealand Palaeodemography
area of 931 pa (a significant 13.8% sample of the 6,956 pa listed in the New Zealand
Archaeological Association’s national site register). Adding the total area for these
931 pa gives an overall total area of almost 4 million m 2 (3,930,000). Multiplying
this number by 6,956/931 gives an estimate for the total area of the 6,956 listed pa
of almost 30 million m 2 (29,361,000). Simple arithmetic enables this estimate for
the total area of New Zealand pa to become the basis for a population estimate for
the major pa areas.
A conservative approach seems appropriate. The first step in such an analysis is
to deal with the issue of double-counting. Large and very large pa seem to have had a
significant role in regional defence, acting as citadels providing defence, especially
against outside attack, not only for local undefended settlements but also probably
for the inhabitants of smaller more vulnerable local pa. From Walton’s data it is
clear that 50% of pa were small with areas of less than 2,100 m 2 . To avoid doublecounting the populations of these smaller pa we suggest discounting the population
capacity of all pa less than 2,100 m 2 in area: that is, we suggest discounting the area
of 50% of all listed pa at the outset. If we discount all small pa, the total area of
the remaining pa reduces from 29,361,000 to 25,660,000 m 2 (a reduction in area
of only 12.6%). As we are hoping to establish the population capacity of pa in
1445, discounting all small pa will also avoid problems arising from the possibility that many small pa may have been built after 1445 to meet the needs of small
groups.
Second, while the building of small pa may have been spread over time, in the
context of rapidly falling populations, we have suggested that it is likely that most
large pa would have been built in the first two generations of the Little Ice Age (AD
1400–1445) when there were large populations to build pa and to defend them. In
Chapter 15 we suggest there has been a confusion in dating some pa where different
stages in a sequence of fortifications have been confused with initial construction.
We note that consecutive disturbances of stratigraphy through a series of sequential
fortifications will have confused dating estimates.
Initially pa may have relied for their defence mostly on the natural defensibility
of the site, with minimal fortifications, perhaps only a scarp and a palisade or a
ditch and palisade, and on having large numbers of warriors to defend them. We
suggest that the sequence of later fortifications adopted for a pa was tied to its “halflife”, that is, tied to each of the stages when, given drastically falling populations,
the numbers of warriors able to defend a pa had fallen to half the number it had
at the beginning of that stage. At the end of this section, we estimate the half-life
of a pa in the Little Ice Age to have been 107.1 years (with 10,000 survivors from
the non-pa areas in 1801) to 108.3 years (with 6,000 survivors). This suggests a
triple sequence of fortifications during the Little Ice Age. This is discussed more
fully in Chapter 15.
In determining the total population capacity of pa in 1445 we have limited consideration to the 90% of large and very large pa that we deem to have been built
by 1445, that is we have limited consideration to 90% of pa which have an area
greater than or equal to 2,100 m 2 . Ninety per cent of 25,660,000 m 2 gives an area
of 23,094,000 m 2 .
14 New Zealand Palaeodemography
area of 931 pa (a significant 13.8% sample of the 6,956 pa listed in the New Zealand
Archaeological Association’s national site register). Adding the total area for these
931 pa gives an overall total area of almost 4 million m 2 (3,930,000). Multiplying
this number by 6,956/931 gives an estimate for the total area of the 6,956 listed pa
of almost 30 million m 2 (29,361,000). Simple arithmetic enables this estimate for
the total area of New Zealand pa to become the basis for a population estimate for
the major pa areas.
A conservative approach seems appropriate. The first step in such an analysis is
to deal with the issue of double-counting. Large and very large pa seem to have had a
significant role in regional defence, acting as citadels providing defence, especially
against outside attack, not only for local undefended settlements but also probably
for the inhabitants of smaller more vulnerable local pa. From Walton’s data it is
clear that 50% of pa were small with areas of less than 2,100 m 2 . To avoid doublecounting the populations of these smaller pa we suggest discounting the population
capacity of all pa less than 2,100 m 2 in area: that is, we suggest discounting the area
of 50% of all listed pa at the outset. If we discount all small pa, the total area of
the remaining pa reduces from 29,361,000 to 25,660,000 m 2 (a reduction in area
of only 12.6%). As we are hoping to establish the population capacity of pa in
1445, discounting all small pa will also avoid problems arising from the possibility that many small pa may have been built after 1445 to meet the needs of small
groups.
Second, while the building of small pa may have been spread over time, in the
context of rapidly falling populations, we have suggested that it is likely that most
large pa would have been built in the first two generations of the Little Ice Age (AD
1400–1445) when there were large populations to build pa and to defend them. In
Chapter 15 we suggest there has been a confusion in dating some pa where different
stages in a sequence of fortifications have been confused with initial construction.
We note that consecutive disturbances of stratigraphy through a series of sequential
fortifications will have confused dating estimates.
Initially pa may have relied for their defence mostly on the natural defensibility
of the site, with minimal fortifications, perhaps only a scarp and a palisade or a
ditch and palisade, and on having large numbers of warriors to defend them. We
suggest that the sequence of later fortifications adopted for a pa was tied to its “halflife”, that is, tied to each of the stages when, given drastically falling populations,
the numbers of warriors able to defend a pa had fallen to half the number it had
at the beginning of that stage. At the end of this section, we estimate the half-life
of a pa in the Little Ice Age to have been 107.1 years (with 10,000 survivors from
the non-pa areas in 1801) to 108.3 years (with 6,000 survivors). This suggests a
triple sequence of fortifications during the Little Ice Age. This is discussed more
fully in Chapter 15.
In determining the total population capacity of pa in 1445 we have limited consideration to the 90% of large and very large pa that we deem to have been built
by 1445, that is we have limited consideration to 90% of pa which have an area
greater than or equal to 2,100 m 2 . Ninety per cent of 25,660,000 m 2 gives an area
of 23,094,000 m 2 .