William Churchill.

Club types of nuclear Polynesia online

. (page 4 of 21)
Online LibraryWilliam ChurchillClub types of nuclear Polynesia → online text (page 4 of 21)
Font size
QR-code for this ebook

ciently close to the 16 pieces of this group to serve as a comparison base.
Between the lengths of 38.5 and 42.25 inches are 7 rootstocks and 9
billets; between the lengths of 43 and 46 inches the record is exactly
reversed — 9 rootstocks and 7 billets. This difference may not be sig-
nificant, yet is worthy of record.

The circumference of the shaft, the first measurements being taken
at the end of the haft, is recorded in table 7.



This set of measurements is essentially comparable throughout the
clubs of whatever type. We shall therefore postpone its consideration
to the later period, when we can discuss it generally upon the basis of
data derived from the whole collection.

The next measurements deal with the circumference of the shaft at
the point where the head-mass has become distinct. This position is

Table 7.

of shaft
(inches) .

Piece No.

of shaft
(inches) .

Piece No.


3303 a
3175. 2483
2481, 3783, 3782 c
2480, 3782 b, 2482


2485. 3782 a
3100, 3782, 2484

not definitely fixed in the structure of the weapon ; therefore there may
be variety of opinion as to the exact spot at which the measurement
should be taken; accordingly the results lack positive value.

These measurements are not properly comparable with those of
table 4, for a new factor has here been introduced. In the billets the
center of gravity affects the shape of the club, which in that case is all
shaft. Here the center of gravity is adjusted to a composite of shaft

Table 8.

of shaft


Piece No.

of shaft

Piece No.

5. 75



3303 a

3782 b, 2482, 3782 a
3175. 3782 c, 3782







2480, 2483, 2479, 2484


and head. Accordingly, the measurements do not mean the same
thing as in the billets ; yet a comparison is profitable as fixing the popu-
larity of certain shapes of club-shaft. Making due allowance for the
difference in the number of pieces in each group, we find that in the
fifth inch the billets are four times as numerous ; in the sixth and sev-
enth they are about even; but in the eighth inch the rootstocks are
four times as numerous. As before, we tabulate, in table 9, the differ-
ence in circumference between the two points of the sh^,ft.

In comparison with the similar statistics of the billets, we find in the
rootstocks a trifle under half and in the billets about the same trifle



over half increasing by an amount between i and 2 inches ; between 2
and 4 inches lie five-eighths of the rootstocks and a scant fifth of the
billets, yet the billets show an increment of from 4 to 5 inches not found
at all in the rootstocks.

Table 9.



Piece No.



Piece No.



3782 a

3783. 3782 b, 3100. 2482, 3782

3782 c, 3303 a, 2479

3 25

3175. 1974. 2485. 2484



The measurements of the length of the head are affected by the same
factor of uncertainty as noted in the last array of figures. Such as
they are, they are presented in table 10.

Table 10.

of head
(inches) .

Piece No.

of head
(inches) .

Piece No.






3303 a
2481. 3783
3782 c
2480. 2485


1974, 2479
2483, 3100, 2482
3782 b, 3782

No significance can attach to these figures; there is in them very
little of the human element of design. A sapHng which above ground
may strike the eye as satisfactory may yield strange results in its root
region, which is to become the head of the club, and these subterranean

Table 1L



of head

Piece No.

of head

Piece No.


(inches) .






3303 a


2483. 2485




2482, 3782 a










3782 b

features are invisible. There is the utmost diversity in the measure-
ments of the heads of these clubs, because they rest upon the irregu-
larities of nature, and the most that man can do is to trim ofif the roots


in a rough fashion. This quahfies the measurements of the circumfer-
ence of the head, as shown in table 11.

All but three pieces (3175. 3783, and 2481) exhibit in varying degrees
the persistence of the descending tap-root already animadverted upon.
In every case it is dealt with independently of the treatment of the
head; its rootlets are cut off flush with the surface and are not dis-
cernible, except upon minute examination of the grain; the surface is
polished; from the raw material it is shaped into a smooth cyHnder or
a cone. Figure 4 in Plate V shows the least trace of art; figure 3 repre-
sents the highest development of the treatment of this element. It
has not seemed advisable to discuss the length of these end-pieces, but
the shape of the end may be noted: domed, 3782 c, 1974, 2483, 2485,
2479; cylinder with end cut square, 3100; conical, 2480.

The haft in this type is characteristically unflanged and cut square,
except that 2479 is domed ; 2484 has a shghtly domed cap; and 3175 (in
this as in most details only temporarily associated with the type) ends
in a flat knob. None of these pieces has perforation, a fact which is
conditioned by the shape and size of the club and equally by the do-
mestic architecture of Fiji and Tonga, from which the type derives.
Any one of these clubs can be stood up in a corner, and the houses of these
two archipelagoes differ from those of Samoa in having side- walls and
resultant corners. Neither has any of them a trace of the lug, and for
the same reason; it is not necessary to suspend such a club when a cor-
ner is within easy reach.

Length, 42.25 inches, of which head is 2.25 inches.

Shaft: Circumference of haft, 4.5 inches; of flat knob, 4.75 inches; at head
7 inches. The plane surfaces of shaft alternating with the
bosses of the head perceptible for 12 inches up shaft. P 3i75-

Head: Conical, with height of 2 inches, surrounded by 6 n?^^^'
bosses 0.25 inch high, of triangular form, i inch on side next piates V i- XV
cone, from which side to opposite apex 1.25 inches; edges and 93, 94, 95, 97.
apices rounded. Each boss carries inlay of cachalot ivory as
follows: (i) figure of eight 0.25 inch high; (2) octopus, 0.75 inch high, point-
ing toward head; (3) triangle, 0.87 inch wide and 0.24 inch high toward head;
(4) octopus, 0.67 inch high toward shaft; (5) triangle, same dimensions as
(3) toward shaft; (6) octopus, as (4) toward head.

Ornament: Grip, 12.25 inches longitudinal band-and-zigzag and basketry
interrupted by broad sinistral spiral band-and-zigzag making 3^^ revolu-

Length, 43 inches, of which head is 3 inches.

Head domed; circular knobs 0.12 inch high, slightly graded in size from
shaft downward, in four tiers, alternating 9 in three tiers and
6 in fourth; shaft molded to head with shallow gutter of 0.12 ^.}A^^-

*°„- ^. r . , , , • , Clark-Oldman.

Haft: Circumference, 4.75 mches; at head, 7.5 mches.

Head: Circumference, 11.5 inches.

Ornament: Grip, 8 inches longitudinal band-and-zigzag.


Length, 43 inches, of which head is 3 inches.

Head: Domed; covered with flat knobs, mostly circular, 0.5 inch diameter,
o. 1 2 inch thick ; 8 pits occupy place of bosses which appear to
have been broken off in use ; three of the pits occupied by plugs 3784 a.
of cachalot ivory, one by human tooth, four empty. ^^^' „

Shaft: Circumference at haft, 4.75 inches; at head, 6.25 piatesII*c-V 2.
inches, flanging head with shallow gutter 0.25 inch.

Head: Circumference, 12 inches.

Ornament: End of haft quartered in panels of radiant lines, shaft to head,
with the exception of 2 panels left blank, covered with longitudinal and
transverse band-and-zigzag and basketry; 2 ties of screw-palm leaf.

Length, 44.5 inches, of which head is 4 inches.

Head: Circumference, 11.5 inches; end highly domed; rootlets slightly
domed; inlay of ivory at end of head; 13 inlays of cachalot
ivory, of which 10 show marks of great age; i root of human 3782 c.
tooth associated with ivory in the same hole ; 4 vacant sockets Pepper-Voy.
designed for inlay; i filled with plug of wood.

Shaft: Circumference of haft, 4.75 inches, at head 7 inches; on end of haft
inlay of cachalot ivory.

Ornament: Grip, 8.5 inches, with 3 parcelings of sennit; stains of sennit
service through rest of shaft; on swelling of head and root portion, 3 lines of
transverse scoring.

Length, 43 inches, of which head is 5 inches.

Head: Circumference, 14 inches; rootlets cut square across with edges
rounded; end of head conical.

Shaft: Circumference of haft, 5 inches; at head 8 inches. ^.i?48o.

Ornament: Grip, 12 inches longitudinal band-and-zigzag, ciark-Oldman.
with median strap of same; one side of head much charred.

Length, 45 inches, of which head is 6 inches.

Head: Circumference, 14 inches; end domed; ends of rootlets carved in
high degree, without sharp edges.

Shaft: Circumference at haft, 6 inches; at head, 8.5 inches. ^..^.974-

Ornament: Grip, 8.75 inches longitudinal band-and-zigzag, Huston collector,
divided by straps of same at 2.5 and 6.5 inches; a final band
slightly indicated.

Length, 45 inches, of which head is 2.5 inches.

Head: Circumference, 10 inches.

Shaft: Circumference at haft, 3.5 inches; at head, 5.75 inches, p-^^®^ *'

Ornament: Grip, 9.25 inches longitudinal band-and-zigzag, with *^'*
initial, median, and terminal straps of same; from terminal strap on opposite
faces of shaft 2 loops of pecked triangular ornament extend for i inch.

Length, 38.5 inches, of w^hich head is 7 inches.

Head: Circumference, 14 inches; end domed; rootlets cut P. 2483 •
square across. -^j*- . ^, ,

Shaft: Circumference at haft, 4.5 inches; at head, 8 inches. ^ ' "^*°'

Ornament: Grip, 10.5 inches service of sennit set on bast foundation,
with I -inch sennit parceling at each end.

Length, 41 inches, of which head is 5 inches.

Head: Circumference, 14.5 inches; rootlets cut square across; end domed.


Shaft: Ciroumfercnce at haft, 5.25 inches; at head, 7.75

inches. P.. 2485-

Ornament: Grip, 10 inches, with service of sennit on bed- p!-''*. qij

ding of bast and longitudinal interlacing of 6 parts of sennit piate VII c.
rounding in loop near end of haft and traces of similar loop
next shaft; next head, 2 twined rings of wire-like rootlets twined in opposite

Length, 41.5 inches, of which head is 8 inches.

Head: Circumference, 13 inches; rootlets cut square across.

Shaft: Circumference at haft, 5 inches; at head, 6.5 inches.

Ornament: Grip, 1 1.5 inches, wrapped with service of sen- 3782 b.
nit, finished top and bottom with parceling of same i inch p •'^' „
long; double strap of glass beads threaded on coir fiber near piate II b V 4.
head. (Compare frontispiece, Seemann's "Mission to Viti.")

Plate V, 5.

Length, 43 inches, of which head is 6 inches.

Head: Circumference 16.5 inches; 10 flanges of considera-
ble regularity, with rounded and smoothed edges ; end lightly ^ .^479-
domed. c\^ k Old

Shaft: Domed; circumference at haft, 5.75 inches; at head, piate V 6.
8 inches.

Ornament: Grip, 8 inches longitudinal band-and-zigzag ; traces of sennit
strap I inch wide next head.

Length, 41.5 inches, of which head is 7 inches.

Head: Circumference, 16 inches; flanges irregular, con-
siderably rounded; end 2 inches high, 2 inches wide, flat. ^.3^°°-

Shaft : Circumference at haft, 5.5 inches ; next head, 7 inches ; p||^''
flattened planes near head alternating with flanges of head.

Ornament: Grip, 10.5 inches longitudinal band-and-zigzag; strap of same
at end and middle; shaft formerly covered with sennit service; natural ring
of bast without seam just below grip.

Length, 42.25 inches, of which head is 7 inches.

Head: Circumference, 15 inches; flanged, with rounded edges.

Shaft: Circumference at haft, 5 inches; at head, 6.5 inches; p..^4o2.

flattened planes near head alternating with flanges. ^ "

Ornament: i inch blank at end of haft, showing stains of sennit ser-
vice; grip, 7.5 inches longitudinal band-and-zigzag, with strap of same;
sennit parceling 1.25 inches; remainder of shaft covered with service
of black and yellow sennit ending next head with 4 ties of pandanus leaf
applied on bed of Turkey-red calico.

Length, 40.5 inches, of which head is 9 inches.

Head: Circumference, 15 inches; narrow flanges with slight rounding; at
distal end of flanges small rootlets sharpened to a point.

Shaft: Circumference at haft, 5.25 inches; at head, 6.5 |.?.?^ *•
inches; next head plane faces merging in roundness of shaft Pepper-Voy.
within 3 inches.

Ornament: Grip, 8.5 inches single and double longitudinal band-and-zigzag;
strap of band-and-zigzag; wTapping of sennit 1.5 inches next end; ornament
ends irregularly and unfinished under wrapping.


Length, 43.75 inches, of which head is 8 inches.

Head: Roots worked in flanges; edges cut square. xr - ^'

Shaft: Circumference at haft, 5.5 inches, at head, 7 inches. Pepner-Voy.

Ornament: Wrapping of sennit, 2 inches; grip, 8.75 inches
longitudinal band-and-zigzag, ending in strap of same; remainder of shaft
parceling of black and yellow sennit for 24 inches, ending in finish of double
strap of sennit in square weaving; head formerly covered with chunam,
abundant traces of which remain in hollows.

Length, 46 inches, of which head is 14 inches.

Head: Rootlets left in natural condition. ^.i^484-

Shaft: Circumference at haft, 5.5 inches; next head, 8 ciark-Oldman.
inches; end of haft slightly domed cap.

Ornament: Grip, 9.75 inches; 2 panels longitudinal band-and-zigzag sep-
arated by single band of zigzag.


Plate I, a, b, c; VIII. Provenience: Fiji.

For reasons of the carpentry of the clubwright it has seemed fit to
introduce at this point the characteristic missile clubs of Fiji. In a
much deeper examination of the source-origin of this club we shall
engage in problems of far greater complexity. But at this point we
are attracted solely by the resemblance of at least one type within this
group to a type frequent among the rootstocks, and by the evi-
dence, both from the illustrations and even more from the observation
of the club-maker at work in Fiji, that all of these missile clubs are
wrought from the sapling and the superior roots quite as much as are
the formidable weapons which we have just passed in review. While
the type is distinctly Fijian, we note its occurrence elsewhere. Kramer
illustrates (210 g) a missile club which appears to correspond exactly
with the ball-head of Plate V, 12, a piece now in Stuttgart accredited to
Samoa ; on the same plate, figure p, he presents from the same museum
and of the same provenience a short club with a shaft rather thicker
than those of the collection here examined, with a ball-head marked off
in small knobs by scorings at right angles, with the final knob of Plate
V, 7, 8, 9, 10, rather more of the type exhibited in the rootstock type
in Plate V. It is quite evident that this club could not be used after
the Fijian method about to be set forth, for the end of the haft is too
large to pierce the tissues. In our vocabulary material we find that
the Tongan has kolo as the name of a club without further definition,
and the Samoan variant 'olo is defined as a short knobbed club carried
by young men, a description which may be taken to apply to Kramer's
210 p.

In the 17 pieces here assembled we may readily discern three posi-
tively distinct types. Of these, that which we may designate the ball
type, 5 pieces, we have the head a more or less regular ball, the shaft of
equal diameter throughout, neither flanging at the haft nor swelling to


the head ; at the juncture of shaft and head a circular depression Uke
a saucer. Structurally this resembles somewhat distinctly the head
treatment of Plate V, 2. Of those in which the head is developed by
the utihzation of the flanges of the stem as in Plate V, 5 and 6, 6 pieces,
and which we may designate the wheel type, we have a shaft which may
or which may not flange at the haft, but which in every case swells into
the head, and at the distal end of the head we find a double element of a
carved ring which may or which may not have a shallow depression
between it and the second element of a more or less carefully worked,
highly domed knob. The third type is distinguished by its carefully
patterned head (Plate V, 7 and 9). In this type the shaft always
swells to the head, and the ring and knob at the head are quite as in the
wheel type. The marking of the head suggests nothing quite so much as
an application of basketry of the pandanus leaf, such as may be seen in
Plate V, 2, as a simple tie about the stem. In two pieces (Plate V, 9)
the flat surfaces of the pattern around the greatest circumference are
distinguished by inner concentric lines, producing the impression of a
double element, of which the lower was allowed to appear through sUts
in the upper.

No matter what the type of this club may be, the Fijian ula, that
affects only its finish, the art of throwing is the same for all and an art
most difficult to acquire. It corresponds generally with the knife-
throwing of the SiciHan and of the mountaineer of Kentucky. In
each case the heavy end of the projectile — the handle of the knife or
the head of the club — goes forward toward the mark. At a certain
distance from the mark the thrown weapon capsizes in a vertical
sense and the blow is delivered by the lighter end — the point of the
knife or the haft end of the ula. The haft end of the club is not
sharpened to a point; examination of the collated measurements will
show it to be of about the thickness of a man's finger or thumb, but
such is its striking force when thrown by the well-muscled Fijian that
it will with no difficulty transfix the softer tissues of the trunk, despite
the protection of the ribs. These ula are thrown with precision to a
distance of 100 feet. In one case I saw a pig pierced through the flank
until the flight of the club was brought up by the head and the haft
came clear through on the other side ; this at a distance of 65 feet.

The club is held pecuHarly, and apparently experience has evolved a
particular grasp and release in order to govern the weapon in its flight.
The haft is held firmly against the ball of the thumb, and it is probably
for this reason that no attempt is made to secure greater piercing
power by fining the haft to a point. The shaft lies upon the half-
flexed middle finger; the index finger slightly bent overHes it; the tip
of the thumb is apphed to keep the shaft in this position; a peculiar
rhythm of the release of the pressure of the three fingers, quite impossi-
ble to describe, seems to be responsible for the distinctive peculiarity



of the flight. Probably the position of the end of the haft, perhaps
some feature of its shape at the point of apposition to the ball of the
thumb, may serve to delay the beginning of the flight and add to the
effect of the rhythmic finger-release. In the 17 pieces of this collec-
tion but one has a domed end; only 5 are cut square across; 11 are
finished at the end with a distinct cupping of the haft, which may be
intended to form a pneumatic joint at the point of apposition, an
opinion which finds support in the additional fact that it is usual to
moisten the ball of the thumb by the tongue before delivering a flight.

Table 12.

(inches) .

Piece No.


Piece No.



2467. 3785 a

2468, 2469, 3785, 3784 a
2461 a, 2461




3786, 2466, 3188
3784, 2465, 2460, 2463
3188 a

The grimness of savage humor is preserved for us in connection with
this weapon ; it may serve to confirm the opinion of those who despise
the pun. The Fijian vakarimbamalamala, meaning literally to cause
chips to fly off, is used to describe a play upon ambiguous words, as
the word ulaula, signifying either to thatch a house or throw short
clubs (ula) at one another. The Bau people sometimes order (we are
quoting from the missionary Hazlewood's Fijian Dictionary, compiled

Table 13.

of shaft
(inches) .

Piece No.

of shaft
(inches) .

Piece No.




2467, 3785 a, 3785





2469, 3784, 2460, 2463, 3784a.

2466, 2461
3786, 2461 a
2465. 3188 a, 3188

at a time when the present tense was accurate) the Tailevu people to
come to Bau to ulaula; the people come expecting to thatch a house and
find themselves pelted with clubs.

The measurements of these pieces are presented in 5 tables, begin-
ning with the length over all in table 12.

For such significance as they may possess, we sum these measure-
ments for each inch of length and its fractions: 15 inches, 3 pieces; 16
inches, 9 pieces; 17 inches, 5 pieces.

The length of the shaft to its point of articulation with the head, a
measurement distinctly obtainable in the ball type but subject to
diversity in the other two types, is presented in table 13.



The circumference of the head is recorded in table 14.

Bearing in mind the highly artful manner of delivering this club
through a hundred feet of air to a precise lodgment upon the mark, it
is clear that there must be some intimate relation existing between
these several measurements of length over all, shaft length, and head
circumference. But the tape-line can not tell all the tale. Before we
can analyze the flight it would be needful to introduce the factor of

Table 14.



of head

Piece No.

of head

Piece No.

(inches) .

(inches) .

9 5

2461 a




2467, 2466




3785 a, 2465



II .25





3785, 2462


3188 a


2468, 3188




3784 a

total weight, the distribution of the weight as between head and shaft,
and the factors of air-resistance conditioned by the volume and form
of the head. In the foregoing tables we note that the length over all
may vary within 2.25 inches, the shaft length within 2 inches, the head
circumference within 9 inches; this record is unmistakably evidential
that the as yet undetermined factors contribute very largely to the

Table 15.

of haft
(inches) .

Piece No.

of haft
(inches) .

Piece No.




2467, 2465, 2462, 2461 a
2469. 3785 a, 3785. 2463, 2466

3 5

2468, 3786, 3188
2460, 3784 a
3784. 3188

art of throwing this missile club. The circumference of the haft is
recorded in table 15.

The end of the haft is cupped in 2468, 2467, 3785 a, 3785, 3784, 2460,
2463, 3784 a, 3188 a, 2461, 3188; domed in 2461 a; cut square in the
remaining pieces.

Perforation is diagonal in 2468, 3785, 3784, 2460, 2463, 2466; and of
these, broken remains of a former perforation are found in 2463, and
in 2466 at least three diagonals, all broken through the septum. In
2461 a we find an unusual perforation, a hole drilled athwart the shaft
I inch from the end and probably the result of modern influence. We
have already noted the elegance of the V-perforation in the center of



the heads of certain larger clubs; in 3785 a we encounter a different
V-perf oration, an inverted V, in which the point is on the head of the
club near the edge and the two limbs appear upon the shaft near the
head. Particular interest attaches to the imperfect inverted V in
3786, for it presents evidence as to the manner in which this perfora-
tion is made — one diagonal is completed, the other has been started
from the side of the shaft, but not yet fully pierced.

Apparently the ring and knob with which the heads of the wheel and
patterned types are finished persists as mere ornament, but it will
complete the record by listing the circumference. These measurements
are probably not comparable inter se; they are to be studied in relation
to the total circumference of the head, even if for no higher purpose
than the study of proportion which satisfies the art sense of the club-

1 2 4 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Online LibraryWilliam ChurchillClub types of nuclear Polynesia → online text (page 4 of 21)