460
Robert L. Welsch
Anthropos 94.1999
The Lewis Collection as Evidence of Cultural
Practices and Social Behavior
The Lewis Collection and other collections of the
period provide evidence (in the form of objects) of
a variety of social behaviors and cultural practices
about which we would otherwise be unaware,
since we cannot today assemble a set of objects
that reliably represents those things actually used
a century ago. Several examples illustrate what the
collection can tell us about the regional distribution
of particular practices.
Several early-twentieth-century ethnographers
had noticed that some Melanesian societies used
the bow and arrow while others did not. Graebner
(1909), among others, saw the bow and arrow as
a cultural survival that could be used as a marker
of past migration routes from Asia out into the
Pacific. While few anthropologists today would
accept Graebner’s view that different kinds of
material culture are reliable as cultural markers or
stable cultural survivals, there is clear evidence in
the Lewis Collection that there was a distinctive
regional variation in most important weapons used
in different parts of the New Guinea coast. Com
munities that used the bow and arrow generally
did not use spears and only a certain cluster
of spear-using villages also used spear throwers.
Thus, bows, spears, and spears with spear throwers
represent three different hunting/fighting strategies
that also happen to be geographically clustered
(see Welsch et al. 1992: Table 2).
Similarly, the clustering of communities that
seem to have used penis gourds (rather than loin
cloths) or sleeping bags to ward off mosquitoes
(rather than other possible strategies) offer other
examples of cultural practices that are geographi
cally clustered, probably because of a mix of
factors relating to intervillage ties and the environ
ment (see Welsch et al. 1992; Welsch and Terrell
1994; Welsch 1996a).
The geographical clustering of these practices
despite of significant linguistic differences sug
gests that some sort of regional processes are at
work, particularly processes analogous to what
Watson (1977) has called the “Jones effect.” The
collection cannot explain such distributions on
its own, of course, but it can provide reliable
historical evidence for distinctive distributions of
different kinds of material culture.
The Lewis Collection as Evidence of Exchange
Of all the cultural practices and social behavior
that are illuminated by the Lewis Collection, the
most significant is what the collection has to say
about trade or exchange.
Because Lewis was concerned with the con
nectedness of different places along the coast, he
paid a great deal of attention to the movement of
objects from village to village. He made comments
in his held notes about what items were exchanged
with other groups and the items that were received
in return. He records information about what items
were valuables with special significance and how
they were made and used. But his notes about
where the different objects he purchased were
made is the single most valuable kind of data he
provides, because it offers empirical data about
the how of a wide variety of valuables about
which almost nothing would now be known from
published sources.
The sheer number of objects (as well as their
variety) that moved from place to place along the
Sepik Coast of New Guinea seems remarkable.
On Ali Island Lewis collected about 317 objects
during a brief visit of only three days in 1909.
But of these Ali specimens, some 81 (25.5 per
cent) were known to have been made elsewhere,
nearly all of these having come from communities
with radically different languages from that spoken
on the island. His collections from the other Ber-
linhafen islands at Aitape (Tumleo, Seleo, and
Angel), are smaller than that from Ali, but show
that all four of these islands shared similar intense
interest in exotic goods from other places along
the coast.
When Tiesler (1969-70) combed more than
200 sources published before 1965 about the Sepik
Coast, he identified some 495 exchange-linkages
between communities, each linkage involving a
specific kind of local product, foodstuff, or raw
material that had moved between two named com
munities. As a single historical resource, the Lewis
Collection contains evidence of some 321 such
exchange-linkages, nearly all of them (97 percent)
being different from those linkages identified by
Tiesler. Of the exchange-linkages cited in Tiesler’s
compilation, only 88 linkages appear in sources
that had been published when Lewis set off for the
held in 1909, so that Lewis’s data provides more
than three times as much evidence of exchange
along the Sepik coast as had previously been
known.
Data in the Lewis Collection shows far more
complex ties between the bush communities (up
to 20 miles inland) and communities on the coast
or islands than is evident in the early literature.
In addition, the Lewis Collection provides tangi