Ornaments of Pandanus leaf. Every married woman always wears a belt of

Pandanus leaf which she is never without for even a moment. When the

belt she is wearing needs renewing she puts the new one on before

taking the other off. To make such a belt two leaves of the Pandanus

Andamanensium are taken and cut to a sufficient length (about 20 cm.).

The thorns at the edge of the leaf are removed by cutting off a strip

of about 3 mm. wide from the edge, leaving a strip of leaf about 4 cm.

broad. The two leaves are placed one on the other and are wound

spirally round in three turns so as to give a belt of six thicknesses

of leaf, the upper surface of the leaf being on the outside of the

belt. The leaves are secured together by tying with thread at the back

where the two ends just overlap. At the point where the leaves are tied

one or two bundles of strips of Pandanus leaf are attached with thread.

The bundle is made of a number of strips of leaf about 90 cm. long and

2·5 cm. broad which are chewed in the mouth to make them soft and then

placed together and served over with thread for about 12 cm. in the

middle. Belts of this kind are generally worn by married women, but

precisely similar belts are worn by men on certain ceremonial

occasions. They are called toto t’er-bua in Aka-Jeru.

A very similar belt is made in exactly the same way save that the

tassel of leaves at the back consists of narrow strips of Pandanus leaf

instead of broad strips. This kind of belt is worn by women only.

Examples may be seen in Plate IV (the lower belt of the two) and in

Plate XVIII. It is called toto t’er-nyarab in Aka-Jeru.

A belt is made in much the same way out of Pandanus leaves split in

half down the midrib, giving strips of about 2·5 cm. broad. Such belts

have only a scanty tassel of thin strips of leaf at the back. They are

worn by girls and women only. They are called kudu in Aka-Jeru.

Yet another variety of belt is made of whole leaves in exactly the same

way as the toto t’er-bua but has no tassel at the back. An example may

be seen in Plate IV (the upper belt). It is worn by women.

Girdles are made of strips of Pandanus leaf of about 1·25 cm. broad

without any tassel, but with strings of Dentalium shell attached at

various points. A girdle of this kind is shown in Plate XV.

Girdles are also made by cutting a number of strips of leaf and

softening them by chewing them in the mouth. These strips are laid

together and either served over or marled with thread so as to make a

girdle of round section. A tassel of leaves similar to that of the toto

t’er-bua is attached to the back, and very frequently strings of

Dentalium shell are attached at various points. Such a girdle may be

worn by either men or women. It is called toto t’er-ŋau in Aka-Jeru.

Yet another kind of girdle is made by splitting Pandanus leaves into

thin strips and making them into a kind of wrapped cord, one strip

being wrapped spirally round one or more others by the same technique

as that used in making bow-strings. A number of coils of strands made

in this way are tied together with thread at various points and a

tassel similar to that of the toto t’er-bua is added at the back. Such

girdles are usually improved by the addition of a few pendent strings

of Dentalium shell. They may be worn by either men or women. Their name

in Aka-Jeru is toto t’er-mo̱i.

Ornaments of Pandanus leaf to be worn either round the leg just below

the knee or round the wrist are made in exactly the same way as the

belt called toto t’er-bua, each having a tassel of Pandanus leaf strips

attached. Such garters and bracelets are worn at a dance by men.

Other ornaments for wearing round the wrist or knee when dancing are

made by the same method as the girdle called toto t’er-ŋau, each having

a tassel of strips of leaf.

Similar dancing ornaments are made by plaiting strips of leaf into a

three-plait cord of the right length, a tassel of loose leaf strips

being left at each end. These are tied round the wrist or knee.

Ornaments for the knee or wrist, such as those shown on the legs of the

man in Plate XII, are made by winding spirally a narrow strip of leaf

and sewing it. A number of pendent strings of Dentalium shell are

attached round it, and often strings are attached at the end of which

small shells hang suspended which rattle against one another as the

wearer walks or dances.

Chaplets or headdresses worn by men when dancing, and occasionally by

women, are made of fairly broad strips of leaf in the same way as the

toto t’er-bua. Such a chaplet may have a tassel of narrow strips of

Pandanus leaf at the back to which are attached shells that rattle as

the wearer moves, or it may have pendent strings of Dentalium shells

round the hinder half of its circumference.

In making ornaments of Pandanus leaf such as the toto t’er-bua the

upper surface of the strip of leaf is usually ornamented with a design

scratched on it with a pointed piece of wood. The design usually covers

not only the portion of leaf that is visible when the belt or other

object is finished, but also that portion which is not visible, being

underneath. Many ornaments of this kind are further decorated when

finished with the composition used for covering the bindings of arrows,

which is applied with a pointed stick in simple geometrical patterns.

The woman shown in Plate X is wearing two belts that have been

ornamented with composition in this way.

Woman’s leaf apron. In the South Andaman the women wear a sort of small

apron consisting of a number of leaves of the Mimusops littoralis laid

one over the other, the stalk ends of the leaves being tucked in

between the layers of a belt of Pandanus leaf. The leaves thus hang

suspended so as to cover, somewhat inadequately, the pudenda. The

natives say that the leaves of the Mimusops are chosen because they

remain green longer than those of any other species. As soon as the

leaves fade and turn yellow they are renewed. The appearance of the

leaf apron may be seen in Plates IV and XV.

The women of the North Andaman have within recent years adopted the

fashion of those of the South Andaman in this matter, but formerly they

made a similar use of a different kind of leaf from a plant called

čainyo in Aka-Jeru, and over the top of the leaves they wore a tassel

of the fibre called ko̱ro.

Ornaments of netting and Dentalium shells. Bands of netting (in shape

of a cylinder or bag open at both ends) are made of suitable size to

tie round the waist, the neck, the leg below the knee or the wrist.

Such netting is generally ornamented as it is being made with strips of

Dendrobium skin worked in the net. Along the lower edge of the band of

netting are attached a number of threads each having Dentalium shells

strung on to it like beads. Ornaments of this kind are shown in Plates

V and IX. They are worn by both men and women, but usually only on the

occasion of a dance or some other ceremony.

Ornaments of bone, etc. These are made with human bones, with the bones

of such animals as pig, turtle, dugong, Paradoxurus, monitor lizard, or

with pieces of wood or cane, or of coral, of suitable shape. A length

of cord, of the fibre of the Hibiscus tiliaceus or of the Ficus

laccifera, is taken, and the bones are attached to it by thread. As a

rule, strips of Dendrobium skin are worked in, being laid on the bone

and bound over with the thread.

Ornaments of shell. Various kinds of shells are perforated and attached

to string and are worn round the neck, the head, the knee, or the

wrist. A necklace of fresh-water shells (ino ko̱lo to̱i in Aka-Jeru) is

shown in Plate IX. A necklace of small sea-shells is shown in Plate

XIII. The favourite shell of the Great Andaman tribes is the Dentalium

octogonum. The shells are collected from the shore by the women. The

closed end of each is bitten off with the teeth and the shells, which

thus form cylindrical beads, are strung on to a piece of thread. These

strings of shell are worn as necklaces, as shown in Plates IV and XII,

and are tied round the wrist and knee and ankle as shown in Plate IV.

Ornaments of seed. The seed-tops of two or three species of mangrove

are collected and strung on to thread and worn round the neck. Fig. 40

shows the two kinds of mangrove seed commonly used in this way.

Bamboo necklaces. A necklace is sometimes made of a number of short

pieces of bamboo arrow-shaft threaded on to a string. The pieces of

bamboo are ornamented with simple designs scratched or cut on them with

a shell.

Sling of bark. Another object that may perhaps be mentioned amongst

objects of personal ornament is the sling used in carrying children. It

is made of a broad strip of the bark of the Hibiscus tiliaceus. Some

slings are covered entirely with netting, while others are ornamented

with shells in various ways. Plate XIV shows one with strings of

Dentalium shell sewn on to it.

Dancing ornaments of Tetranthera wood. A piece of Tetranthera wood,

usually part of the shaft of an old pig-arrow, about 30 to 35 cm. long,

is taken and made into shavings with a Cyrena shell. The wood is shaved

carefully round and round, so as to make a continuous sheet of fibre,

as though unwinding a roll of material. A bundle of these shavings is

tied at one end and covered with red paint, and forms an object that is

carried in the hand or worn in the belt at dances.

PERSONAL ORNAMENT; LITTLE ANDAMAN.

The personal ornaments of the Little Andaman and of the Jarawa are

different from those of the Great Andaman Division, and therefore need

to be described separately.

Ornaments of bark. Strips of bark (? Celtis cinnamonea) are worn by the

men round the waist and round the arm.

Ornamental cord. These are the ornaments most frequently met with both

in the Little Andaman and amongst the Jarawa. The basis is a strip of

cane, varying in breadth in different examples. On one side of this are

laid strips of the yellow skin of the Dendrobium, varying in number

according to the breadth of the cane, and the whole is served over or

bound with thread. The technique in shown in Fig. 41, which represents

diagrammatically the method adopted when there are two strips of

Dendrobium skin. Such ornamental cord is made in pieces of considerable

length. Often tassels of thread (of Ficus hispida fibre) are attached

to it at intervals. The cord is worn wound round the waist, the neck,

or the arm. Both men and women wear it.

Woman’s girdle. In the Little Andaman this is made of a number of fine

strips of cane tied together with thread. At the front is attached a

tassel of fibre made from young unopened palm-leaves (the fibre called

ko̱ro in Aka-Jeru).

Woman’s shoulder strap. The women of the Little Andaman wear a strip of

bark over the shoulders, crossing over the chest and passing beneath

the breasts.

The above brief description of the ornaments of the Andamanese is

perhaps sufficient for the purpose of this appendix, which is to

determine as far as possible the elements of a primitive Negrito

culture. Any complete account of the subject would need a large number

of figures and a discussion of many comparatively unimportant details.

The first point of importance to be noted is that the personal

ornaments of the Little Andaman Division differ markedly from those of

the Great Andaman Division. One difference is that the Pandanus leaf,

which is used so much and in so many ways in the Great Andaman, seems

not to be used at all in the Little Andaman. Another is that shells,

which are much used in the Great Andaman, are used to a far less extent

in the Little Andaman. The natives of the Little Andaman told me that

they sometimes make ornaments of Dentalium shell, but I did not see any

such ornament, nor any other ornament of shell, in use. The ornaments

of netting worn in the Great Andaman, and forming the usual dancing

costume, seem not to be used in the Little Andaman. Finally, perhaps

most important of all, there is at present no evidence that the natives

of the Little Andaman make ornaments of either human or animal bones.

In the Little Andaman the lower jaw of a child is preserved by the

parents and is worn by them, but I never saw a human skull (whether of

child or adult) so worn, and I was not able to obtain any evidence of

the use of strings of human bones such as are constantly seen in a camp

of the Great Andaman.

The points of similarity between the ornaments of the Little Andaman

and those of the Great Andaman are very few. One of the most striking

is the great use that is made in both divisions of the yellow skin of

the Dendrobium. It seems probable that here we have evidence of one

element of a primitive Negrito culture, for the Negritos of the

Philippines also are fond of using a yellow vegetable fibre for their

personal ornaments. So far as it is possible to judge from the figure

and description, the armlet given by Meyer in Plate iii, No. 11, seems

to be very similar to the ornamental cord of the Little Andaman. It is

described as a “mit schwarz und gelben Grasstreifchen umflocktener

Bambus-reif.”

Another point of resemblance between the Great Andaman and the Little

Andaman is that in both very little use is made of feathers or flowers.

The natives of the Great Andaman never make use of the feathers of

birds. The Jarawa occasionally wear in a chaplet a feather of the

king-crow (Dicrurus macrocercus) if we may judge from a specimen in the

British Museum. The natives of the Little Andaman certainly do not as a

rule make any use of feathers. It would seem, from all accounts, that

the Semang and the Philippine Negritos do not make any considerable use

of feathers for personal ornament. We may perhaps hazard the conclusion

that this is a mark of the Negrito culture distinguishing them from

such people as the Papuans or Australians in whose personal ornaments

the feathers of birds occupy an important place.

In the Great Andaman flowers are not used as personal ornaments. In the

British Museum there is an ornament from the Little Andaman consisting

of a strip of bark-fibre with a few flowers attached.

As regards personal ornament, therefore, the only elements of a

primitive Negrito culture that we seem to be able to trace in the

Andamans at the present day are (1) the use of yellow vegetable fibre,

and (2) the absence of any considerable use of feathers.

One element of a primitive Andamanese culture, though not necessarily

of a primitive Negrito culture, would seem to be the use of a tassel of

the fibre obtained from an unopened palm-leaf which is used by women to

cover their genitals in the Little Andaman, and was similarly used

until recently in the North Andaman.

HAIR-DRESSING, SCARIFICATION, AND BODY-PAINTING.

At the present time the usual method of hair-dressing in the Great

Andaman is to shave a portion of the scalp all round so as to leave a

sort of skull-cap of hair, as may be seen in many of the plates of this

volume. In some cases a “parting” is made by shaving a narrow strip

over the crown. (See Plate XV, for example.) When the hair grows so

long as to be uncomfortable the whole head is shaved, and it is then

permitted to grow again. In these days the natives cut their hair

whenever they have an opportunity of obtaining a pair of scissors. In

the Little Andaman women and old men are frequently to be seen with the

head entirely shaved. The younger men shave away the lower edge of the

hair all round in the same way as the natives of the Great Andaman, but

this may be a recent practice.

The fashion of hair-dressing, at any rate for women, has changed within

recent times, for Mr Man wrote in 1882 that “the majority of the women

every week or ten days shave their heads almost entirely, leaving only

two narrow parallel lines of hair, termed gor, from the crown to the

nape of the neck [256].” At the present time this style of

hair-dressing has fallen entirely into disuse, and the women do their

hair in the same way as the men.

The operation of shaving, which is done with a flake of glass or

quartz, is performed by women, and never, or very rarely, by men.

Mention has been made in an earlier part of the work of the way in

which the natives of the Great Andaman scarify the skin of the body and

limbs with a flake of glass or quartz. The Semang do not scarify

themselves in this way, but some, at any rate, of the Negritos of the

Philippines do.

The natives of the Great Andaman, as described earlier in this work,

paint their bodies with a grey clay called odu or og, with a fine white

clay, and with red paint made by mixing burnt oxide of iron with fat or

oil. The natives of the Little Andaman use the same kind of clay as

that called odu or og in the Great Andaman, but instead of applying it

in patterns they smear it roughly on the back and front of the trunk.

They also use red paint, with which they smear their hair, a practice

never met with in the Great Andaman.

ORNAMENTATION.

The Andamanese have simple designs which they paint or incise on their

bodies and on a great number of the objects that they make and use. A

few typical designs such as are incised or painted on belts of Pandanus

leaf are shown in Fig. 42. A very large number of designs are based on

the zig-zag line. Examples are shown in Fig. 42 c, e, and g. In all

parts of the Great Andaman and also in the Little Andaman the zig-zag

line is associated with snakes. Thus in the Little Andaman the simple

zig-zag line is called dobo kwolage (dobo = snake). In the North

Andaman the design of zig-zag lines painted on the body with white clay

is called or-čubi t’era-bat, or-čubi being the name of a species of

large snake.

A number of other common designs consist of parallel lines of dots or

of short strokes, an example of which is shown in Fig. 42 f.

By far the greater number of the Andamanese designs are based on the

following elements, (1) parallel lines crossing a surface from side to

side at right angles to the edge or else in a sloping direction (about

45°), (2) parallel lines of dots or of short lines, i.e. parallel

broken lines, as in Fig. 42 f, (3) zig-zag lines, which may be single,

or parallel or opposed so as to make lines of lozenges. As an example

of the way in which these elements may be combined two designs copied

from bamboo necklaces are shown in Fig. 43.

It would perhaps be possible to show that there is a real connection

between the ornamentation of the Andamanese and that of the Semang, as

there is certainly a considerable degree of superficial resemblance,

but at present we understand so little the psychological processes

underlying the use of ornament amongst primitive peoples that the

subject is one of considerable difficulty.

CANOES.

Canoes are in regular use on the coast in all parts of the islands.

There are three types of canoe. (1) The Little Andaman canoe, with one

outrigger, propelled with paddles or with a pole. (2) The Great Andaman

small canoe, with one outrigger, propelled with paddles or with a pole.

(3) The Great Andaman large canoe, without outrigger, propelled with

oars.

The third kind of canoe mentioned above is a recent innovation. The

natives themselves say that such canoes have only been made in recent

times, since they have been able to obtain a plentiful supply of iron

tools, and so have been able to cut down and hollow out large trunks

[257]. They seem to have been invented by the natives of the South

Andaman, and copied by those of the Middle and North Andaman.

Canoes of this type are propelled by rowing with short oars, except in

shallow water where they are poled. This method of propulsion (rowing

as opposed to paddling) was adopted in imitation of the boats with

which they have become familiar since the European occupation.

For the small canoe of the Great Andaman five or six species of

soft-wooded trees are used, of which three are species of Sterculia. A

suitable tree is selected near the shore or a creek, and is felled.

Care is taken to make it fall in a particular direction. Thus, if the

trunk is curved, the convex side of the curve will have to be the

bottom of the canoe, and the tree should fall so that this side lies on

the ground. Trees are very rarely regular and before beginning the work

of cutting the natives have to decide how it should be cut so as to

give the best result, i.e., the greatest stability.

After the tree is felled the trunk is cut to the requisite length. The

inside of the canoe is first roughly hollowed out with the adze, no use

being made of fire. The bark is then removed from the outside of the

trunk and the two ends are shaped. Finally, the inside is carefully

finished with the adze so as to reduce the sides and bottom to the

requisite thickness.

Except at the stem and stern, the canoe retains the shape of the tree,

only the bark being removed, and the sides and bottom being formed of

the alburnum or sap-wood. At the stern a small platform is cut

projecting over the water, which serves as a seat for the steersman. At

the prow a larger platform is cut, on which the harpooner stands when

he is harpooning turtle or fish. Below these two platforms the ends are

not cut away squarely but are rounded from side to side. The prow of

the canoe is in every case the lower and therefore broader end of the

trunk. It is only in this way that a sufficiently large platform can be

provided for the harpooner.

The trees used for canoes have a pithy core and there is therefore a

small patch in both the prow and the stern which would admit the water.

In former times these two places were caulked with bees’-wax. At the

present time the natives often nail a piece of tin (part of an old

kerosene tin, for example), with some rags beneath it, on the outside

of the canoe at these two places.

When the hull of the canoe is finished it is moved to the shore or to

the bank of a creek and the outrigger is attached. The float is a

straight spar of light wood. In the North Andaman the wood of the

Hibiscus tiliaceus is often used, but Mr Man says that in the South

Andaman the float is always made from a species of Sterculia (mai in

Aka-Bea) [258]. The ends of the float are only roughly shaped. The

broader end of the float is forward.

The float is attached to a number of booms, of which there are never

less than three in the smallest canoes, while there may be as many as

eight or nine in a large canoe. A medium-sized canoe has five or six

booms. The boom is a thin straight piece of tough wood, of which one

end is sharpened and thrust right through two holes cut in the gunwales

of the canoe opposite to one another, the sharpened end projecting for

a few inches on the port side of the canoe. The boom thus projects

about three feet on the starboard side on the level of the gunwale.

Where the boom passes through the gunwales of the canoe it is bound

with cane, and the cane is bound round the whole of that part of the

boom that is within the canoe between the two gunwales. (See Fig. 44.)

This portion of the boom forms a seat for the man paddling the canoe,

so that he sits on a level with the gunwale with his feet on the floor

of the canoe.

The boom is attached to the float by means of sticks of tough wood.

These sticks, having pointed ends, are driven into the float, one

perpendicular, and the other two at an angle on each side. The tops of

the three sticks are fastened to the boom a few inches from its end by

means of a strip of cane. The arrangement of the three sticks is shown

in Fig. 45.

The strip of cane with which the sticks are bound to the boom is wound

spirally round the boom itself for a few inches and is then carried

down round the float and back to the boom again on the other side. The

three sticks provide an efficient resistance against a longitudinal

thrust (i.e., a thrust in the same direction as the line of the canoe).

The strip of cane passing from the boom round the float and back to the

boom again provides a resistance against any lateral thrust on the

float. The three sticks, being driven in when the wood of the float is

dry, do not readily work loose, as the water in which it is constantly

immersed keeps the wood swollen. The cane binding, including the stays

on each side, may work loose, but can readily be tightened or renewed.

Each of the booms is attached to the float in exactly the same way, and

the whole arrangement is very efficient in keeping the float rigidly

attached to the hull in such a position that it rests on the surface of

the water when the hull itself rides freely balanced.

Canoes of this type vary in dimensions within wide limits. A small

canoe with only three booms, which would carry three persons, measured

4·85 metres in length over all with a beam of about 35 centimetres. A

large canoe may measure as much as 9 metres with a proportionate beam.

A well-made canoe will often balance well enough as it stands, but it

is sometimes necessary to balance it with ballast of stones or pieces

of coral. In any case the canoe is easily overturned in a rough sea

unless the occupants can maintain the balance with their bodies. As the

canoe is made of light wood it cannot sink even when full of water, and

the natives easily right an overturned canoe, bale it out, and get in

again, even in a rough sea.

The furniture of a canoe consists of the ballast (of stone), a piece of

stone (or sometimes a piece of tin) on which to keep a small fire

smouldering, an anchor consisting of a lump of coral or stone attached

to a length of rope, a Nautilus shell or two for baling out the canoe,

a bamboo pole of about 18 feet in length for poling the canoe in

shallow water, and paddles.

In the platform overhanging the prow a few holes are cut. These holes

are sometimes used to attach the rope by which the canoe is fastened to

an anchor. One or more turtle skulls are often attached so as to hang

down beneath the platform. In turtle-hunting expeditions on dark nights

a torch is slung beneath the forward platform so as to shed its light

on the water while the harpooner remains in shadow.

When a canoe is finished it is decorated with designs painted on it

with red paint and white clay, particularly on the forward platform and

along the gunwale. These designs soon wear off when the canoe is in use

and are not renewed.

The paddles used by the Andamanese vary considerably in size and to

some extent in shape, but the following description with its

accompanying figure gives a fair idea of a typical specimen. The whole

paddle has a length of 123·5 centimetres, the shaft being 85

centimetres and the blade 38·5 centimetres. The diameter of the shaft

from the blade to the middle is 2·6 centimetres, and from the middle

towards the handle end it tapers to a point. The shaft is circular in

section throughout its length. The blade is leaf-shaped, pointed

bluntly at the apex. In section it is plano-convex, with a maximum

thickness of 1·3 centimetres and a width of 8·7 centimetres. Paddles

are cut with an adze from the wood of the Myristica longifolia and

planed with a boar’s tusk. They are often ornamented, when new, with

painted designs in red and white.

In deep water the canoe is paddled. Each of the occupants sits facing

forward. The steersman sits on the stern platform. The others sit on

the seats provided by the outrigger booms. Each man paddles on which

side he chooses. It rests with the steersman to maintain the canoe in

its proper course. In shallow water the canoe is propelled with a pole.

A man stands on the forward platform and poles the canoe, steering as

he does so. In a fairly large canoe a man at the stern may also take a

pole and, standing up, help to propel the canoe.

All the work connected with the making of canoes and paddles is done by

men alone, except the painting, which is usually done by women. It is

the men also who make most use of canoes.

The large canoe, that is now made by all the tribes of the Great

Andaman Division, is simply a canoe of the same general shape as the

Great Andaman outrigger canoe, cut from a larger tree and without the

outrigger. The shape of the hull, with its platforms fore and aft, is

exactly the same. It seems that when the natives obtained a plentiful

supply of iron tools (after 1858) they began to cut down and hollow out

larger trees than formerly. Having made these larger canoes they found

that they would, when well cut and ballasted with stone, float quite

well without an outrigger. (It may be mentioned that the hull of a

small canoe is always tested on the water before the outrigger is

attached.) Indeed a well cut canoe of large size floats and balances

better in a rough sea than a smaller one with an outrigger. It is

possible that at first these large canoes were propelled with paddles

just as the outrigger canoes are, the paddlers facing forward. Having

learnt to understand the principle of the oar, through their contact

with the Penal Settlement, the natives applied this principle to their

own canoes. It could not, of course, be applied to the small canoes, as

the shape of the canoe and the position of the paddler make the use of

an oar impossible. It could be applied very easily, however, to the new

large canoes. In these the oarsman does not sit on a level with the

gunwale, but sits down in the hull itself on a piece of wood resting on

the two sides of the hull a few inches above the floor. The gunwale of

the canoe is thus about on a level with the bottom of his sternum. A

number of holes are made in the gunwale on each side, and by means of

these, loops of cane are attached to the gunwale. The oar, which is

shaped in imitation of European oars, but with a short shaft, is thrust

through the loop of cane, which serves as a fulcrum or rowlock. The

rowers face aft. A man at the stern steers with a paddle. In shallow

water the large canoes are propelled with poles in exactly the same way

as the smaller canoes.

In the bow, at about the position that would be occupied by the

foremost boom in an outrigger canoe, holes are made in the gunwale on

each side and a piece of wood is thrust through them as a sort of

thwart. This is to provide a means of making fast the end of the

harpoon line or the anchor line, and thus serves a purpose that is

served by the foremost boom in the outrigger canoe.

The large canoes are not quite so useful in turtle-hunting as the

smaller outrigger variety, as they cannot be so quickly turned when the

pursued turtle doubles. Very often a large canoe and a small canoe are

taken together on an expedition, the harpooning being mostly done from

the smaller one while the captured turtles are placed in the larger

one. A small canoe with three or four men cannot hold more than one or

at most two big turtles, whereas as many as ten or a dozen can be

stowed in one of the large canoes.

The chief use of the large canoes is to make journeys from place to

place. One of the largest will hold as many as thirty men and women

with their baggage, whereas an outrigger canoe would never carry more

than nine or ten. Further, there is less chance of an even heavily

laden big canoe capsizing in a rough sea than of an outrigger canoe

doing so. One result of the introduction of the large canoe has

therefore been to enable the natives to move much more freely about the

islands than formerly. The passage from the South Andaman to Ritchie’s

Archipelago, for instance, would only be attempted in an outrigger

canoe on a very calm day, whereas in a large canoe it can be

successfully accomplished even when there is something of a sea

running.

The small canoe of the Little Andaman is fairly similar to that of the

Great Andaman. There are three differences. (1) The stem and stern are

squarely cut in the Little Andaman, instead of being rounded off. (2)

The outrigger booms are attached to the top of the gunwale by cane

binding which passes through holes made in the gunwales, instead of

being themselves passed through holes in the gunwale. (3) The float is

attached to the booms in a different and less efficient manner. Three

pointed hard-wood sticks are driven into the float, but they are all

three approximately perpendicular. They are bound at the top to the

boom, but there are no stays of cane to maintain the float rigid

against a lateral thrust. In other respects the Little Andaman canoe is

the same as the Great Andaman canoe.

The Jarawa of the South Andaman do not at the present time make use of

canoes. This is apparently because, through their hostility with the

Aka-Bea, they have been confined to the interior of the island. They

make rafts of bamboos lashed together for crossing creeks and inlets.

The forest-dwellers of the Great Andaman Division seem also to have

made occasional use of similar rafts for the same purpose.

A canoe of the North Sentinel was seen by Mr Gilbert Rogers during a

visit to that island in 1903 [259]. It had been hollowed out of a tree

and was about 15 feet long. The ends of the canoe were cut off

perpendicularly to its length leaving a piece of the tree about one

inch thick projecting for about three inches beyond either end to form

a small but rickety seat. The log from which the canoe was cut was

curved so that the ends were slightly higher than the middle. It had a

float supported by six booms passing through holes cut in the sides of

the canoe. These booms were fastened to smaller pieces of stick fixed

into the sides of the canoe beneath them. The outrigger was attached to

each boom by two small pointed sticks driven into the float and tied to

the boom above with cane. There was one small paddle, a Nautilus shell

for a baler, and five poles resting on the outrigger booms. These point

to the canoe being poled along in the lagoon, which is quite shallow,

rather than to its being used for long journeys or outside the reefs

which surround the island. The canoe was 18 inches in diameter at the

smaller end and perhaps 30 inches wide at the larger end.

This description shows that the canoe of the North Sentinel is on the

whole more like that of the Little Andaman than like that of the Great

Andaman.

It seems probable that the Andaman Islands were peopled by sea from the

coast of Burma. If this were so, then the original ancestors of the

Andamanese must have been in possession of canoes. A consideration of

the present Andaman canoes suggests that their ancestors had canoes

with a single outrigger on the starboard side, with a number of booms.

Of the different methods of attaching the booms to the float, it is

possible that the method now in use in the Little Andaman (and

apparently also in the North Sentinel) is primitive, and that the Great

Andaman attachment is an improvement that has been invented since the

separation of the two divisions. On the other hand it is not impossible

that the Great Andaman attachment is primitive, and that in the Little

Andaman we have a degeneration that might be due to the fact that the

Little Andaman (and equally the North Sentinel) provides much less

scope for maritime pursuits than the Great Andaman.

The recent invention of the large canoe in the Great Andaman and the

adoption of the principle of the oar shows that the Andamanese readily

adopt new inventions when these are clearly of service to them.

Turning now to the Semang, as these people live inland they have no use

for canoes. They make rafts of bamboos lashed together with which they

float down the rivers, returning overland.

Some of the Philippine Negritos seem to live on the coast and possibly

have canoes, but nothing is known about these.

CONCLUSION.

The examination of the technical culture of the Andamanese given above

has been sufficient to enable us to make a few statements as to what

was probably the culture of the Negritos before they were split up into

isolated groups. It is highly probable that they obtained their

subsistence solely by hunting and collecting vegetable products. They

had bows and arrows, the form of bow being probably fairly similar to

that used at the present day in the Little Andaman, while for hunting

the larger animals they had arrows with detachable heads. They possibly

had no knowledge of any way of making implements of stone, but made use

only of such materials as wood, bone, and shell. It is not probable

that they possessed the art of making pottery, and their basketry and

mat-work were probably confined to very simple forms. In their personal

ornaments there is reason to think that they showed a fondness for

bright yellow vegetable fibre, and made little or no use of the

feathers of birds. The ornamentation of their utensils was probably

confined to the use of the simplest forms of geometric design with a

preponderance of the zig-zag and the lozenge. Finally their huts

consisted of a single sloping roof sufficient to afford shelter for a

single family or larger huts consisting of such small huts joined

together.

We have seen that since the Andamanese have occupied their present

home, or at any rate since the Great Andaman Division and the Little

Andaman Division have been separated from one another, many changes,

some of considerable importance, have taken place in the Andaman

technology. In general it would seem that the technical culture of the

Great Andaman has changed more than that of the Little Andaman. Putting

aside the effect on the technology of the introduction of iron, there

is no evidence that any of the changes that have taken place in the

Great Andaman have been due to outside influence. Important

modifications have taken place in the form of the bow, in the forms and

technique of baskets, and in personal ornaments, and in all these

instances there is no reason to think that these changes have not been

brought about by the natives themselves without the influence of

contact with other people. Their method of working iron, based as it

is, to all appearance, on their former method of working shell, shows

that even here, though the iron itself came to them from outside, and

even though they may have learnt its use from seeing it used by aliens,

still they have not learnt from others how to fashion the metal into

shape by heating it. Thus, so far as their technical culture is

concerned, there is no evidence whatever that the Andamanese have ever

been influenced by contact with any other race since the time, now many

centuries ago, when they first reached the islands.

On the other hand there is some probability that the ancestors of the

Andamanese, before they first reached the islands, or at any rate

before the isolation of the Little Andaman from the Great Andaman, had

learnt from some other race how to make pottery, and it is possible

that at the same time they may have acquired other elements of their

culture, such as the outrigger canoe. We may even give a guess as to

the particular culture from which the ancestors of the Andamanese may

have adopted these elements, which may well have been that of a branch

of that people of whom an offshoot peopled the Nicobars.

Confirmation of these hypotheses, if confirmation be ever forthcoming,

can only be obtained in the study of the history of races and of

culture in south-eastern Asia. Until we have much fuller knowledge of

the culture of the Semang and the Negritos of the Philippines, any

conclusions that may be drawn from the study of the Andamanese alone

must be regarded as provisional working hypotheses only, and it is as

such that they are here put forward.

APPENDIX B

THE SPELLING OF ANDAMANESE WORDS

In writing words of the Andaman languages I have used a slightly

modified form of the “Anthropos” Alphabet of Father Schmidt [260],

which I consider to be by far the most scientific alphabet for writing

down the languages of primitive peoples. The consonants are

k g y ŋ

č ǰ

ń

t d n l r

p b w m

The letter ŋ is used for the nasalised guttural stop (ng in English)

which should always be written with one letter, since it is a single

consonant, quite distinct from the double consonant ng of “ungodly.”

The letter ń stands for a palatalised n, something like the sound in

French “agneau.” The č and ǰ, which, in the “Anthropos” Alphabet

represent the sounds in English “church” and “judge” respectively,

should, I think, really be written t′ and d′. The t′ is a palatalised

t, as heard in “Tuesday,” whereas the č is a fricative, often regarded

as a compound of t and sh. It is not always easy to distinguish t′ from

č and d′ from ǰ, but I believe the Andamanese sounds are really t′ and

d′, and this is to some extent confirmed by the fact that they have no

s, z, sh or zh in their languages. I have used the č and ǰ because

former writers had written these sounds ch and j and it seemed worth

while to make some sacrifice of scientific exactness in order to avoid

too great a divergence in spelling from previous workers in the same

field.

The remaining consonants may be pronounced as in English. I have not

distinguished between different varieties of the consonants l, r, t, d,

k and g. Further I have not distinguished between p and p̌ (the labial

fricative). Many of the words of the Northern languages that I have

written with a p are often pronounced with a p̌ sound.

The vowels are

i u

e o ö

e̱ o̱

a̤ a̱̱

a

These may be pronounced as follows:

i, intermediate between the vowels of “it” and “eat.”

e, as the vowel in “say.”

e̱, as the e in “error” or the a in “Mary.”

a̤, as the a in “man.”

a, as the a in French “pas.”

a̱̱, as the a in “path.”

o̱, as the vowel in “not” or in “nought.”

o, as in “go.”

u, as the vowel in “fool.”

ö, nearly as the German ö.

I have not attempted to distinguish all the different varieties of

vowel sounds that are found in the different dialects. Slightly

different but closely related sounds are represented by the same letter

[261].

In writing Andamanese words I have followed the practice of separating

by hyphens the affixes from the stems in each word. In the Andamanese

languages there are two main classes of words. The first class consists

of words each of which is a simple stem (without affixes). Such words

are the names of what the Andaman Islander regards as simple

independent objects or things, such as roa, canoe, baraba, a mat, ra, a

pig. The second class consists of words each of which is formed of a

stem and a prefix. In the language of the Little Andaman (and in that

of the Jarawa as far as known) these prefixes are simple vowels, i-,

e-, a-, o-, and u-. In the languages of the Great Andaman Division they

are such as ot-, aka-, era-, e-, u-, ab-, etc. Such words are used to

denote dependent objects such as the parts, qualities or actions of a

thing. Thus while a pig (ra) or a canoe (roa) is an independent object

and is therefore denoted by a simple stem, the head or prow of a pig or

canoe, being a part of a “thing,” is denoted by a word (ot-čo)

consisting of a stem (-čo) and a prefix (ot-). So also the quality of

bigness which may belong to a pig or a canoe is denoted by a word

(er-kuro) consisting of a stem (-kuro) and a prefix (er-).

Thus the second class includes many words that we should call nouns,

all the words we should call adjectives and practically all those we

should regard as verbs. In the Andamanese languages the distinction

between nouns and adjectives is not very clear and even less so is the

distinction between adjectives and verbs. Whereas the distinction

between things (independent objects) on the one hand, and the parts,

qualities, actions, etc. of things on the other, is of the utmost

importance. For this reason I have thought it advisable always to

separate by a hyphen the prefix and the stem in words of the second

class.

In compound words or phrases in which the second word consists of

prefix and stem it is common to insert before the prefix a t or an l.

Thus ra t’er-kuro is “a big pig” and ra t’ot-čo is “a pig’s head.” In

writing such words I have placed an inverted comma between the t or l

and the prefix. This infix t must not be confused with the first

personal pronoun in the North Andaman, Tio, contracted to T’ before a

prefix.

NOTES

[1] “Notes on the Languages of the Andaman Islands,” by A. R. Brown,

Anthropos, Vol. IX, 1914, pp. 36–52 with map. This paper contains notes

on I, The Relations of the Andamanese Languages, and II, The Formation

of Words in the Language of the Little Andaman.

[2] I hope to be able to publish shortly the first volume of a work in

which the same method is applied to the interpretation of the social

institutions of the natives of Australia.

[3] The formation of the Arakan Fold (including the Andaman and Nicobar

Islands), dates from the middle of the Tertiary Period, and was

apparently connected with the great movements that produced the

Himalaya-Alpine mountain system and the Circum-Pacific Fold. The

Andaman Sea, in the later Tertiary period, was prolonged much further

to the north, over the region now occupied by the Pegu Yomah.

[4] This line of volcanic activity is a minor continuation of the Sunda

Range of volcanoes of Java and Sumatra. It is continued northward,

parallel to the Arakan Fold, as far as the extinct volcano of

Puppadoung, east of Pagan, not far from Lat. 21°.

[5] The flora of the Andamans and Cocos contains a number of species,

the presence of which can only be explained by the supposition of a

past land connection with the Arakan region. (See Prain, “The

Vegetation of the Coco Group,” Journ. Asiatic Soc. of Bengal, Vol. LX,

Part II, pp. 283–406.) On the other hand, the paucity of mammalian

fauna is such as to lead to the conclusion that the islands were

isolated at a period when the mammals now typical of the mainland did

not exist there. (See Miller, “Mammals of the Andaman and Nicobar

Islands,” Proc. National Museum, U.S.A. Vol. XXIV).

[6] There is no evidence of the former existence of Negritos in the

Nicobars, but on the other hand, there is equally no direct evidence of

their former presence in Lower Burma.

[7] On the accompanying map of south-eastern Asia the regions now

occupied by the Negritos are shown by the shading.

[8] Anciennes Relations des Indes et de la Chine; De Deux Voyageurs

Mahometans, qui y allerent dans le neuvième siècle; Traduites d’Arabe

(par M. l’Abbé Renaudot). Paris, MDCCXVIII, pp. 5 and 6.

[9] Takakasu’s Edition of I-tsing, pp. xxviii seq.

[10] The Travels of Marco Polo, Edited by John Masefield, Everyman’s

Library, 1908, p. 347.

[11] Extracts of Master Caesar Frederike: his Eighteene Yeeres’ Indian

Observations, Purchas: his Pilgrimes, London, 1625; Vol. II, p. 1710.

In spite of the repeated descriptions of the Andamanese by early

writers as ferocious cannibals, there is good reason to think that they

have not deserved quite so evil a reputation. If they had ever been

cannibals they had certainly abandoned the custom by the time the

islands were occupied in 1858. It is improbable that such inveterate

man-eaters, as they are supposed to have been, would have entirely

altered their ways in the course of a century or two. The legend

probably had its origin in the fact that the Andamanese attacked all

strangers who landed on their coasts, and (in the North Andaman, at any

rate) often disposed of the bodies of slain enemies by cutting them in

pieces and burning them on a fire.

[12] The account is that of a visit to the Andamans in 1771 by John

Ritchie, published in the Indian Antiquary, Vol. XXX, 1901, pp. 232

seq. Two natives came off to the ship in a canoe, and Ritchie writes:

“I gave them some nails and bits of old iron which pleased them much;

and about three in the afternoon, they went into the canoe, and tried

hard to pull the chain plates from the vessel’s side. They went astern

when this would not do, and dragged strongly and long at the rudder

chains; but these were too well fixed; and at last, they went towards

the shore at an easy rate, looking at their nails, and singing all the

way.”

[13] In 1906 some Little Andaman visitors to Rutland Island captured a

Jarawa of that part. They told me that though he spoke differently from

them, they could understand him fairly well.

[14] Portman, M. V., A History of Our Relations with the Andamanese,

Calcutta, 1899.

[15] This estimate is based on what the Andamanese were able to tell me

of the conditions under which they formerly lived. Of course such an

estimate can only be of small value. I think it is more probable that I

have underestimated the former population than that I have

overestimated it.

[16] Notes on the Languages of the South Andaman Group of Tribes, p.

27.

[17] Page 18 above.

[18] A few small areas were not occupied at all, for example the

greater part of Saddle Peak in the North Andaman, which is covered with

dense jungle and is supposed by the natives to be the haunt of large

and deadly snakes and of evil spirits.

[19] Mouat, F. J., Adventures and Researches among the Andaman

Islanders, London, 1863, p. 313.

[20] Loc. cit. p. 300.

[21] Journ. Anthrop. Inst. Vol. XII, pp. 107 and 108.

[22] Portman, Notes on the Languages of the South Andaman Group of

Tribes, p. 23.

[23] The photograph is reproduced in Le Tour du Monde, 1895, p. 447.

[24] See Journ. Anthrop. Inst. Vol. XII, p. 71.

[25] Man, op. cit. p. 108.

[26] In the North Andaman the times before the Settlement are spoken of

as the time when there were no dogs, Bibi poiye = “Dog not.”

[27] They are only eaten in the rainy season.

[28] There are two kinds of wild bee in the Andamans. A small species

makes black honeycombs in hollow trees, and these may be found at any

time of the year. A larger species of bee builds white combs suspended

from the underside of branches in tall trees. Such combs are found in

abundance only in the hot season, and not at all in the middle of the

rainy season.

[29] Man, in Journ. Anthrop. Inst. Vol. XII, p. 108.

[30] The natives of the North Andaman were able to tell me of a few

cases of murder which had occurred within the memory of those still

living.

Mr Portman in his History of Our Relations with the Andamanese records

a certain number of murders which occurred while he was in charge of

the Andamanese. One man, who had been imprisoned at Port Blair for

murder, committed another soon after his release and was hanged. Since

that date there has been no case of murder among the Great Andaman

tribes. This is perhaps in part due to the punishment with which they

are now threatened by the Government, but another cause is probably the

breakdown of the old social organisation which has in this respect

rather improved their morals than the opposite.

[31] The nickname is applied, however, not only to those who deserve it

by their character, but also to others; for instance, one man was

called Tarenǰek because his maternal uncle was a man of violent temper.

[32] See below, Chap. IV.

[33] The classificatory system of relationship was first studied and

named by Lewis H. Morgan, in Systems of Consanguinity and Affinity of

the Human Family, Washington, Smithsonian Institution, 1871. The

subject is also discussed in the same author’s Ancient Society.

Although there has been a good deal of attention paid to the systems of

relationship of savage tribes since the time of Morgan, there is no

general work on the subject that supersedes these two books.

[34] The terms used in any society to denote relationships are of

interest to the ethnologist as an important means to the discovery of

the relationship system (i.e. the system of juridical and moral

institutions) existing in the same society. Without a thorough

knowledge of the terms in use and their exact meanings it is impossible

to discover the rights and duties of relatives one to another. It is,

however, sometimes forgotten that the study of terms of relationship is

not an end in itself but a means to a more important study.

[35] In the Andamanese languages a large number of words are formed

from a stem and a prefix. E-, ot-, aka-, ara-, ab- etc. are prefixes of

this kind. The function of the prefixes is (1) to show that the object

denoted by the word is in a dependent relation to some other object

understood, as for instance that it is part of that other object, and

(2) to modify the reference of the stem, as for instance while e-tire

means the offspring of an animal or an human being, era-tire means the

offspring of a tree or plant (the young shoots). For a description of

these prefixes the reader may be referred to the work of Mr Portman,

Notes on the Languages of the South Andaman Group of Tribes.

[36] Dege bula and dege pal mean “my husband” and “my wife”

respectively.

[37] Notes on the Languages of the South Andaman Group of Tribes.

[38] Man, op. cit. p. 421. The dia, or the d’ before a prefix, in the

words of this list is the pronoun “my.”

[39] Portman, Notes on the Languages of the South Andaman Group of

Tribes, p. 255.

[40] Op. cit. p. 100.

[41] The natives commonly applied the term to me, in the form Mam-jula.

[42] The systems of relationship of savage peoples are often very

difficult to study, even with a thorough mastery of the native

language. My account of the Andamanese system is not perhaps complete

and is therefore open to error. Since the above account was written I

have had the opportunity of studying in Australia several examples of

“classificatory” systems of relationship, and can now say very

definitely that such a system presents an extreme contrast to the

system of the Andamans. My failure fully to comprehend the Andamanese

system was partly due to the difficulties of the language, in which I

did not have time to become expert, and partly to the nature of the

Andamanese terms, of which it is by no means easy to discover the

meaning, even with careful observation.

[43] Man, op. cit. p. 135. He speaks of the wives as “models of

constancy.”

[44] I collected a number of genealogies from the natives, but

unfortunately my own inexperience in the use of the genealogical

method, and my consequent inability to surmount the difficulties with

which I met, made this branch of my investigations a failure.

[45] When a husband and wife greet one another the man sits on the lap

of the wife.

[46] Man, op. cit. p. 139.

[47] Ibid.

[48] Man, op. cit. p. 139.

[49] Man, op. cit. p. 125.

[50] It would not be safe, however, to base any arguments of importance

to sociology on the above description of the Andamanese system of

relationship alone. Although I tried to learn all that I could on the

subject, it is quite certain that I did not learn all that was to be

learnt, and it is possible that further enquiry might have shown that I

was mistaken in some of my observations. The difficulty of being really

sure on these matters is due (1) to the fact that the breaking-up of

the old local organisation has produced many changes in their customs,

and (2) to the difficulty of questioning the natives on matters

connected with relationships when they have no words in their language

to denote any but the simplest relationships.

[51] The dance is described in the next chapter.

[52] In the years 1872 to 1902 inclusive the Jarawa made eight attacks

on camps of the friendly Andamanese in different places, in which two

of the friendly Andamanese men and one girl were killed and three men

and one boy were wounded. There were also one or two casual meetings

between Jarawa and friendly Andamanese. One of the friendlies was

surprised and killed while turtle hunting in 1894. During the same

years the Jarawa made on different occasions about twenty attacks on

parties of convicts or on separate individuals, killing altogether 27

convicts and two police constables, and wounding six other convicts. In

these skirmishes and in the expeditions to which they gave rise three

Jarawa were killed and seven wounded on various occasions, and several

times Jarawa men, women or children were captured and afterwards

released. A number of convicts have at different times run away from

the Settlement and as some of those were never after heard of they may

be supposed to have been killed by the Jarawa. For an official record

of dealings with the Jarawa see the “Census Report” 1901, pp. 68–90.

[53] I could not obtain any definite legend about these stones, but one

informant said that when Biliku got angry and destroyed the world (see

later, Chap. IV) the children all became stones at this place.

[54] This is the plant (not identified) of which the leaves were, till

recent times, worn by the women of the North Andaman to cover the

pudenda. In the South Andaman the leaves of the Mimusops littoralis are

in use for this purpose, and the Northern tribes have recently given up

their own custom and adopted that of the South.

[55] Unfortunately I was not able to see this ceremony performed, and

my information is therefore derived from the statements of the natives.

[56] There is no secrecy about any of the proceedings; the whole

ceremony is performed in the village and may be witnessed by anybody.

[57] In the Southern tribes large stones are placed on the youth’s

thighs.

[58] In these days the natives are very fond of tea, which they obtain

from the Andamanese Homes; during the ceremony described above the

youth or girl is not permitted to drink tea.

[59] I believe that the dance is intended to imitate the movement of a

turtle as it swims through the water.

[60] The meaning of the word kimil (or gumul) will be discussed in a

later chapter.

[61] This plant is selected because it is associated with

honey-gathering; its bitter sap, being extremely obnoxious to bees, is

smeared over their persons when taking a comb, and enables them to

escape scot free with their prize. (Note by Mr Man.)

[62] This is believed to be the Ophiophagus elaps. (Note by Mr Man.)

[63] Man, op. cit. p. 133.

[64] Man, loc. cit. p. 143.

[65] Man, op. cit. p. 354. Mr Man adds in a note that “it is believed

that Puluga would punish severely any person who might be guilty of

eating his yat-tub, either by causing his skin to peel off (wainyake)

or by turning his hair white and flaying him alive.” On Puluga see

later, Chaps. III and IV.

[66] Turtle are captured alive by means of harpoons, and may be kept

alive several days before they are killed and eaten.

[67] Mr M. V. Portman gives a list of personal names in use in the

South Andaman in his Notes on the Languages of the South Andaman Group

of Tribes, p. 70. The derivations of many of the names as there given,

are, however, of doubtful accuracy.

[68] Portman, Notes on the Languages of the South Andaman Group of

Tribes, pp. 166–188.

[69] A similar custom is found in many savage tribes. Thus in many

parts of Australia the aborigines call white men by the same name that

they apply to the spirits of the dead.

[70] Man, op. cit. pp. 158, 159.

[71] Portman, Notes on the Languages, etc. p. 183.

[72] Man, op. cit. p. 152.

[73] Ibid. p. 160.

[74] Man, op. cit. p. 160.

[75] Ibid. p. 161.

[76] Portman, Notes on the Languages of the South Andaman, p. 328.

[77] Portman, Notes on the Languages of the South Andaman, p. 314.

[78] Man, op. cit. pp. 153, 154.

[79] Man, op. cit. p. 165.

[80] Ibid. p. 153.

[81] Man, op. cit. p. 157.

[82] Man, op. cit. p. 112.

[83] Man, op. cit. p. 159.

[84] Ibid. p. 158.

[85] Man, op. cit. p. 153.

[86] Man, op. cit. p. 159.

[87] Man, op. cit. p. 161.

[88] I could not obtain any information about the word that Mr Man

gives as chaitan. Some men of the South Andaman whom I questioned did

not seem to recognize the word, except as their way of pronouncing the

Urdu word shaitan = devil.

[89] In the “Census Report” 1901, p. 62, Sir Richard Temple writes,

“The Andamanese have an idea that the ‘soul’ will go under the earth by

an aerial bridge after death, but there is no heaven nor hell nor any

idea of a corporeal resurrection in a religious sense.”

[90] It may be noted that in the Andamanese languages there is no

future tense of the verb, and it is often very difficult to know

whether a speaker is referring to the present or to the future.

Further, although there is a past tense, a native often uses the

present tense in a narrative relating to the past, so that a statement

relating to the past and one relating to the future may have exactly

the same grammatical form. Mr Ellis, writing in the Journal of the

Philological Society (1882) from information supplied by Mr Man, gives

a verbal suffix -ngabo denoting the future in the Aka-Bea language. Mr

Portman points out that this is an error. (Notes on the Languages of

the South Andaman, p. 88.)

[91] Man, op. cit. p. 96.

[92] Man, op. cit. p. 83.

[93] In preparing the fibre, the skin or bark of the young shoots of

the plant is torn off in strips and these are placed on the thigh and

scraped with a Cyrena shell.

[94] Man, op. cit. pp. 153 and 172.

[95] Portman, Notes on the Languages, etc. p. 97.

[96] The legend will be given later.

[97] Man, op. cit. p. 169.

[98] The suffix -la is added to personal names and to terms of address

in order to express respect.

[99] The name seems to mean “alone.”

[100] The giant bamboo does not grow in the Andamans, but pieces of it

are often drifted ashore, having come from the coast of Burma. The

natives pick up these drift-wood bamboos and make buckets of them. It

is possible that the bamboo from which the first man was born was just

such a piece drifted up from the sea. Unfortunately I neglected to

enquire on this point when taking down the legend.

[101] The meaning of the name is “the cave of Tarai”; I believe that

this is the name of a spot in the Aka-J̌eru country.

[102] The meaning of the name was not discovered.

[103] The lizard was caught in some way by his genital organs, but I

was unable to understand the story completely.

[104] This is the small platform of sticks placed near or above the

fire, on which the natives keep their food, and on which they often

place objects that they desire to dry.

[105] Notes on the Languages, etc. p. 227.

[106] When an old man of the A-Pučikwar tribe was giving me the

information repeated above, an Andamanese man was with us who had been

brought up as a Christian and had some knowledge of the doctrines of

that faith. He explained to me that Tomo is the equivalent of the

Christian God. This man belonged to the Akar-Bale tribe.

[107] These names are common personal names among the aborigines of the

present day. Mr Portman derives Nyali from nam-da, the name of a tree,

and Irap from pira-da meaning “scattered,” but these derivations are

far from being authenticated. (Portman, Notes on the Languages of the

South Andaman Group of Tribes, p. 70.)

[108] The place called Irap is at the north end of Havelock Island.

[109] Man, op. cit. p. 164.

[110] Until the settlement of Europeans on the islands the Andamanese

had no knowledge of any means of producing fire. It is necessary to

remember this to understand some of their legends which relate how in

the time of the ancestors the fire was very nearly lost in a heavy

storm.

[111] I understood that Lirtit, by the loss of his wings and tail,

became a man.

[112] Portman, loc. cit.

[113] Mo̱m is a title indicating respect, and Mirit is the imperial

pigeon.

[114] Portman, loc. cit.

[115] Ibid.

[116] Portman, loc. cit.

[117] Man, op. cit. p. 164.

[118] Ibid.

[119] Man, op. cit. p. 167.

[120] Dik was one of the ancestors. He was a giant and was so big that

he could go into the deepest water and never needed a canoe. He used to

shoot dugong and porpoise with his bow and arrow. (The natives shoot

small fish with a bow and arrows, but large fish and dugong and

porpoise they take with harpoons.)

[121] Karami is the name of a bird that was not identified.

[122] Kočurag-boa is the Akar-Bale name for a huge legendary animal.

[123] When a man has killed another, either in a personal or a tribal

quarrel, he has to observe several customs of which one is to keep

himself painted with red paint for several weeks.

[124] Man, op. cit. p. 171.

[125] Man, op. cit. pp. 167–169.

[126] Notes on the Languages, etc. p. 27.

[127] Man, op. cit. p. 166.

[128] Ibid.

[129] Man, op. cit. p. 169.

[130] This is the name of some creature that I did not identify,

perhaps a kind of spider.

[131] An Andaman Islander will often, when walking along the shore,

shoot his arrows before him, either aiming at some object, or trying to

send each one as far as possible. I have never seen them do this in the

jungle, for they might easily lose the arrows.

[132] The Andamanese classify resin as a “stone” although they know its

vegetable origin.

[133] Man, op. cit. p. 172.

[134] Kalwadi is a small crab, yaramurud is the crow pheasant

(Centropus andamanensis), and toau is the hawksbill turtle.

[135] Knives are generally carried slipped into a string that is tied

round the neck, the knife, with a skewer of sharpened wood that is

attached to it, hanging at the back of the neck, where it is easily

accessible and not likely to get lost.

[136] I could obtain no explanation of the phrase, or word, to̱ kit. My

informant only said “That is the way the spirits talk.”

[137] Man, op. cit. p. 165.

[138] The sneezing (the word is translated literally) is a sort of

whistling noise that the wild pigs make when they suspect danger.

[139] The Andamanese always disembowel a pig and sever the joints of

its legs before they place it on a fire.

[140] Man, op. cit. p. 170.

[141] Page 214.

[142] Man, op. cit. p. 170.

[143] In taking a honeycomb the natives often drive away the bees with

fire or smoke.

[144] In climbing a tall tree the Andamanese choose a stout cane or

other creeper depending from one of the branches of the tree, and climb

up it.

[145] The natives express their joy at a success in hunting by

shouting.

[146] Man, op. cit. p. 173.

[147] The narrator said “resin.” The Dipterocarpus tree does not

produce resin, but a sort of oil. The marks on the two fishes owe their

origin to this incident.

[148] Man, op. cit. p. 173.

[149] Man, op. cit. p. 169.

[150] Vide supra, p. 142.

[151] The making of such hypothetical reconstructions of the past has

been regarded by a number of writers as the principal if not the sole

task of ethnology. My own view is that such studies can never be of any

great scientific value. Although, within narrow limits, particularly

when the method is applied to the facts of language and material

culture, it is possible to reach conclusions of some degree of

probability, yet by their very nature all such hypotheses are incapable

of verification. Moreover, the purpose of scientific studies is to

discover general laws, and hypotheses as to events in the past of which

we have and can have no certain knowledge will not provide suitable

material from which to draw generalisations.

[152] It may be worth while to mention that the interpretation of

Andamanese customs given in this chapter was not worked out until after

I had left the islands. Had it been otherwise I should have made

careful enquiries into subjects which, as it was, escaped my notice.

[153] Tylor, Primitive Culture, I, 387.

[154] Frazer, Totemism and Exogamy, IV.

[155] Max Müller, Physical Religion, p. 119.

[156] Marett, Threshold of Religion.

[157] McDougall, Introduction to Social Psychology, Chap. XIII, seems

to combine the two hypotheses.

[158] For a criticism of the hypotheses of animism and naturism as

explanations of primitive religion see Durkheim, Les Formes

Élémentaires de la Vie Religieuse, Book I, chapters 2 and 3.

[159] Sentiment,—an organised system of emotional tendencies centred

about some object.

[160] See p. 73 above.

[161] Page 134.

[162] In a few words the psycho-physical theory here assumed is that

weeping is a substitute for motor activity when the kinetic system of

the body (motor centres, thyroid, suprarenals, etc.) is stimulated but

no effective action in direct response to the stimulus is possible at

the moment. When a sentiment is stimulated and action to which it might

lead is frustrated, the resultant emotional state is usually painful,

and hence weeping is commonly associated with painful states.

[163] McDougall, Social Psychology.

[164] Active sympathy, the habitual sharing of joyful and painful

emotions, is of the utmost importance in the formation of sentiments of

personal attachment.

[165] It is a commonplace of psychology that a collective emotion, i.e.

one felt and expressed at the same moment by a number of persons, is

felt much more intensely than an unshared emotion of the same kind.

[166] It will be shown later in the chapter that when individuals are

excluded from participation in the dance it is because they are in a

condition of partial exclusion from the common life.

[167] The psychology of dancing offers a wide field for study that has

as yet, so far as I know, been barely touched. The following brief

notes are therefore necessarily incomplete and somewhat unsatisfactory.

[168] I have known a dance to be continued for seven or eight hours,

each dancer taking only short periods of rest; and it must be

remembered that the Andamanese dance is more strenuous than our

ball-room dances.

[169] Man, op. cit. p. 76.

[170] Man, op. cit. p. 333.

[171] In a number of tribes of Western Australia I found an exactly

similar custom. It was formerly the rule that after the death of a near

relative the mourner must abstain from eating kangaroo, that being the

largest game animal. Since the establishment of sheep stations in their

country, with the consequent great decrease in numbers of the kangaroo,

it has come about that the animal which now provides their most

important supply of meat is the sheep, and the modern rule is that a

mourner must not eat mutton.

[172] Page 171.

[173] The brakes formed by the cane (bido) from the leaves of which the

ko̱ro fibre is obtained seem to be regarded as lurking places of the

spirits. The natives often speak of the Bido-teč-lau (Calamus leaf

spirits).

[174] Page 216.

[175] The psychological function of individual anger is to restore to

their normal condition the wounded self-regarding sentiments. The

function of collective anger is similarly to restore the collective

sentiments on which the solidarity of the society depends.

[176] I once drew a few grotesque figures for the amusement of some

Andamanese children, and they at once pronounced them to be “spirits.”

[177] The Andamanese beliefs about storms and the weather generally

will be dealt with in the next chapter.

[178] See above, p. 119.

[179] See p. 227.

[180] I am unfortunately obliged to leave a big gap in this chapter and

in the book, owing to my inability to discuss the Andamanese notions

about sex. The natives of the Great Andaman at the present time show an

unusual prudery in their conversation and dealings with white men, but

there is good reason to suspect that this is due to the influence of

officers who have been in charge of the Andaman Home in former years.

At the present time all the men except a few of the oldest in remote

parts are very careful never to appear before a white man without some

covering although formerly they wore nothing. In their conversation in

the presence of a white man they are careful to avoid reference to

sexual matters. The men of the Little Andaman who have not come under

the influence of the Andamanese Homes, still go naked and unashamed,

and indulge in obscene gestures and jokes. At the time I was in the

Andamans I failed to realise the very great importance of a thorough

knowledge of the notions of a primitive people on matters of sex in any

attempt to understand their customs, and therefore failed to make the

necessary enquiries.

[181] In order to carry the analysis further it would be necessary to

consider in detail the whole question of the relation of art and

ceremonial, and that of the social function of art which is involved in

it, and also to deal with the notion of “value” as it appears in

primitive societies. The material from the Andaman Islands is not

suitable for the discussion of these problems.

[182] The exposition of this important thesis can only be given here in

the most abbreviated form. The thesis itself, as applied to primitive

ritual in general, owes its origin to Professor Émile Durkheim, and has

been expounded by him (more particularly in his work Les Formes

élémentaires de la Vie religieuse) and by Messieurs H. Hubert and M.

Mauss.

[183] Primus in orbe deos fecit timor.

[184] Page 214.

[185] We have seen, in the last chapter, that any condition of the

individual in which he is withdrawn from active participation in the

common life is regarded as one of danger from magico-religious forces

antagonistic to the society.

[186] It will be shown later in the chapter that some part of the

respect paid to the cicada is due to its connection not with the day

and night but with the seasons of the year.

[187] Page 215.

[188] Page 221.

[189] Page 294.

[190] Page 215.

[191] Page 213.

[192] Pages 207 and 204.

[193] Page 207.

[194] Ibid.

[195] Page 204.

[196] Page 203.

[197] Page 207.

[198] Page 202.

[199] Page 206.

[200] Ibid.

[201] Page 207.

[202] Page 208.

[203] The same threefold division of the world is seen in the beliefs

about the three kinds of spirits, those of the forest, those of the

sea, and the Morua who, while spoken of as spirits of the sky, are

often thought of as living in the tops of the tall trees.

[204] Page 218.

[205] It is worth while to recall here the belief that if a man goes

into the water after eating civet-cat he will not be able to swim.

[206] Page 208.

[207] Ibid.

[208] Page 209.

[209] Page 206.

[210] Page 207.

[211] It appears also in geographical names. Puluga-l’ar-mugu, meaning

‘the Puluga front’ is the name of a part of the Archipelago facing the

N.E. and means ‘the side facing Puluga.’

[212] Although it is generally believed that storms (or more exactly,

violent storms or cyclones) are the results of the anger of Biliku, yet

there is a conflicting belief that storms are made by the spirits,

particularly the spirits of the sea.

[213] Page 154.

[214] The application of the name biliku to the spider is clearly a

minor motive, and probably a late accretion. The name of the N.E.

monsoon is the same in all the divisions of the Andamans about which we

have information, with dialectic differences only. In the Little

Andaman the form of the name is Öluga, and the same name is given to

the monitor lizard. Presumably, therefore, there was originally one

name throughout the Andamans for the N.E. monsoon (Öluga, Puluga,

Bilik, Bilika, Biliku) and later this name was applied to the spider in

the North Andaman and to the monitor lizard in the Little Andaman. It

may be noted that the name of the monitor lizard varies from one

language to another in the Great Andaman.

[215] It is to be noted that these tabus connected with Biliku are not

absolute prohibitions; they are beliefs that if certain things are done

Biliku will be angry (i.e., there may be storms); if you do these

things you must risk the danger. It is exactly the same with the

roasting of pork.

[216] See, for instance, the Aka-J̌eru legend on pages 197–198, the

Aka-Kede on page 200 and that from the Akar-Bale tribe on pages

200–201, and also the legends on pages 207, 208.

[217] In a paper in Folk-lore, vol. xx, 1909, I put forward the

hypothesis that probably at one time all the tribes of the Andamans

regarded Biliku (Puluga) as female, and Tarai (Daria) as male. I am

still inclined to think that there is some evidence for this, but a

discussion of what the Andamanese beliefs may have been in the past is

entirely outside the scope of this chapter and is therefore omitted.

[218] The stem be seems to be connected with the idea of cutting.

[219] In dealing with the account given by Mr Man of the Andaman

mythology it is necessary to remember that he was undoubtedly

influenced by a very strong desire to show that the beliefs of the

Andamanese about Puluga were really fundamentally the same as the

beliefs of the Christian about his God. It may be taken as certain that

he did not consciously allow this wish to affect his record of the

Andaman beliefs, but it is very improbable that it did not

unconsciously have a great deal of influence both on Mr Man and on his

informants.

[220] To complete the discussion of this part of the subject it would

be necessary to deal with many points in the legends of the real

meaning of which I do not feel satisfied. I have, for instance, given

no explanation of the position of Pe̱rǰido in the Biliku-Tarai myth,

although this is probably an important matter. Nor have I traced to its

source the connection of Biliku (with her net, and her hole, or cave,

in which she shuts herself up to sleep and from which she comes out to

bring rain and storm) with the spider. Besides Tomo, Biliku has yet

another competitor for the position of control over the fine weather of

the hot season, namely the snake, or-čubi (wara-ǰobo), which is

regarded as being in some way the guardian of honey and of fine

weather. There are legends that show the connection of this snake with

honey (page 227) and the same connection is shown in the honey-eating

ceremony (page 105). According to Mr Man, when the natives of the South

Andaman see a dark cloud approaching and they do not wish it to rain

they threaten Puluga that they will call up the wara-ǰobo to bite him.

The snake, like other snakes, is only to be seen during the hot weather

of the honey season. It may be remembered that it is from this snake

that the pattern used in decorating the body with white clay is named.

[221] In the last chapter it was shown that the attribution of magical

force to such things as foods and human bones is simply the means by

which the social values of these things are represented and recognized.

Similarly here the magical powers of the ancestors are simply the

representation of their social value, i.e. of the social value of

tradition.

[222] In Central Australia it is believed that if a boy who has not

been initiated eats large lizards he will develop an abnormal and

diseased craving for sexual intercourse. (Spencer and Gillen, Native

Tribes of Central Australia, p. 471.) A friend who has observed the

monitor lizard in Australia tells me that the animal fully deserves its

reputation.

[223] Page 217.

[224] Page 218.

[225] Page 218.

[226] See Introduction, p. 12.

[227] The information here given as to the Semang is derived from two

works, Skeat and Blagden, Pagan Races of the Malay Peninsula, vol. 1,

quoted as Skeat, and Rudolf Martin, Die Inlandstämme der Malayischen

Halbinsel, 1905, quoted as Martin.

[228] The information about the Philippine Negritos is derived from

Reed, Negritos of Zambales, Manila, 1904, quoted as Reed, and A. B.

Meyer, Die Philippinen, II, Die Negritos, Dresden, 1893, quoted as

Meyer.

[229] See p. 34.

[230] See Census Report, 1901.

[231] Supplement to the Andaman and Nicobar Gazette, January 2, 1904.

[232] Skeat, p. 172.

[233] Skeat, p. 174.

[234] Skeat, p. 176, and plate. See also Annandale, Fasciculi

Malayensis, Anthropology, Part I, Plate IV.

[235] Skeat, p. 174.

[236] Skeat, p. 177.

[237] Reed, Plate xxxviii.

[238] Meyer, Plate x.

[239] Skeat, p. 341.

[240] Skeat, p. 270.

[241] Skeat, p. 280.

[242] Skeat, p. 205.

[243] Skeat, p. 205.

[244] Reed, p. 44.

[245] Reed, p. 44.

[246] Reed, p. 47.

[247] Reed, p. 47.

[248] Reed, p. 48.

[249] For information about the two specimens at Cambridge I am

indebted to the kindness of Mr J. W. Layard.

[250] According to Skeat the end with the longer point is the lower end

of the bow (Skeat, p. 273), but Martin (p. 785) describes a bow of this

type and states that the longer point is the upper end.

[251] Meyer, pp. 13–17 and Plates VI–VIII.

[252] Meyer, Plates VI and VIII.

[253] Skeat, p. 274.

[254] Lapicque, “Ethnographie des Iles Andaman,” Bulletin de la Société

d’Anthropologie de Paris, 1894, p. 370.

[255] Skeat, p. 383.

[256] Man, op. cit. p. 77.

[257] Man, Journ. Anthr. Inst. XII, p. 367, note 4.

[258] Man, op. cit. p. 397.

[259] Supplement to the Andaman and Nicobar Gazette, January 2, 1904.

[260] Published in Vol. ii of the journal Anthropos, 1907.

[261] Although I had acquired some knowledge of phonetics before I went

to the Andamans, as a necessary part of the preliminary training of an

ethnologist, yet it was not really sufficient to enable me to deal in a

thoroughly scientific manner with the problems of Andamanese phonetics,

and my further studies of the subject give me reason to believe that my

phonetic analysis of the Andaman languages was not as thorough as it

might have been.