-~7

L I B R.AFLY
OF THE
UN IVERSITY •
Of ILLI NOI5

5*72.05

FA
v. 35 -3"?

M

5TZJ

FA

y.36

LDIANA • ANTHROPOLOGY

Published by
CHICAGO NATURAL HISTORY MUSEUM

Volume 36 September 19, 1945 Number 1

POTTERY FROM THE ALEUTIAN ISLANDS

George L Quimby

Curator of Exhibits, Department of Anthropology

The aboriginal inhabitants of Amaknak Island (Dutch Harbor,
Alaska) were the Eskimo-Indians called "Aleut" by the Russians.
Before the arrival of white men, they occupied all of the Aleutian
Islands and part of the Alaska Peninsula.

The prehistoric Aleut made their living by hunting sea mammals,
fishing, and gathering wild plants and shellfish. They lived in
underground houses, dressed themselves in garments of fur, skin,
or gut, traveled in skin-covered boats, and made tools and weapons
of bone, stone, and driftwood. They also wove dried grasses into
baskets and mats. The social structure of Aleut society was rela-
tively more elaborate than that of their distant kin, the Eskimo
of northern Alaska.

Aleut culture was modified by the arrival of Russian explorers and
traders in the middle of the eighteenth century. Subsequent exploi-
tation of the natives by white hunters and traders nearly extin-
guished the fire of Aleut culture and the Aleut as well. In a little
more than fifty years the population decreased from somewhere
between 16,000 and 25,000 to fewer than 2,000. However, Aleut
culture can be reconstructed from the records of early explorers,
hunters, and traders, and the accounts of later missionaries and
archaeologists.

According to historical records and the investigations of early
archaeologists the Aleut did not make pottery, although they some-
times added clay walls to their stone frying pans.

Recently, however, pottery has been discovered in the Aleut
midden sites on Amaknak Island (Dutch Harbor). The discovery
was made by Lieutenant A. R. Cahn, U.S.N.R., who salvaged con-
siderable material from sites or parts of sites destroyed in the course
of military construction. Lieutenant Cahn presented a large col-
lection of Aleut artifacts (including pottery) to Chicago Natural
History Museum. With a few minor exceptions, all of the pottery

No. 571 1

THE LIBRARY OF THE

NOV 7 1C45

2 FIELDIANA: ANTHROPOLOGY, VOLUME 36

came from Midden C. The minor exceptions consist of two sherds
from Midden A, one from Midden D, and two from Midden G.
In all there were about thirty sherds, some of them representing
sizable parts of vessels, from Midden C. Some of these sherds were
from 6 to possibly 13 feet beneath the top of the midden.

Stone vessels (fig. 4, c-f) which closely resemble those made of
pottery were also found in the middens, particularly in Midden C.
Much of the pottery is so coarse, thick, and heavily tempered with
sand or rock fragments that it can only with difficulty be separated
from the stone. Because of this similarity as well as the similarity
of vessel forms it is probable that earlier investigators encountered
pottery in the Aleutians, but mistook it for stone.

All of the pottery is crude, very thick, and heavy. Probably it
was made by modeling the clay into the desired shape. Tempering
is granular in character and always exceptionally abundant; other-
wise it varies. Some of the pottery seems to be heavily tempered
with sand, or possibly a very sandy clay was used. Most of it, how-
ever, was tempered abundantly with gritty particles of various
sizes. Several fragments of this pottery (representing two different
vessels) were sent to Dr. Anna Shepard for examination. She says
of them: "They are composed of volcanic material — fresh sharp
mineral grains and fragments of a basalt — weakly bound by a red
clay. Not only does the nature of the material indicate that they
are pottery, but also the characteristics of the surface, for I find
finger-like depressions in one area and a smooth flat surface with
coarse particles of temper imbedded in it. This latter effect would
be produced had the paste been pressed against a smooth flat surface
but could not have been obtained had the vessel been worked with a
tool, for the material is exceedingly friable and the weak binding of
the clay would not hold the particles in place to be evenly worked
to the level of the surface. I did not take time for specific petro-
graphic determination of the inclusion, because the question was
only whether or not the vessels are pottery."

None of the pottery is decorated. With the exception of a small
lamp all of the vessels seem to have been elliptical or oval bowls
with flat bottoms and slightly convex sides. The walls range in
thickness from 2 to 6 cm. Generally the vessel bottoms are thinner
than the sides (1-4 cm.). The following are some selected examples
of Aleut pottery in the Museum collection.

One restored vessel (fig. 1), found at a depth of 63^ feet in the
C Midden, is tempered with sand. I would guess that there is as

FA
\t.3U1.

Fig. 1. Pottery vessel which has been mended and restored.

4 FIELDIANA: ANTHROPOLOGY, VOLUME 36

much sand as clay in the paste of this pot. The pottery is buff on
the inside and smoke-blackened on the outside. The vessel is ellipti-
cal, with a flat bottom, a vertical, slightly convex rim, and a rounded
lip. The pot is 33 cm. long, 25 cm. wide, 10 cm. high, and 3 cm. thick.

A large fragment (fig. 2, a) from the C Midden probably was
part of a rather shallow bowl with thick walls (3 to 3.5 cm.), a
slightly outslanting convex rim, and a rounded lip. The paste is
abundantly tempered with small angular particles of grit and with
sand. Of all the sherds in the collection this one is the most easily
recognizable as pottery.

A small black sherd (fig. 2, c) from Midden C is abundantly
tempered with grit particles ranging in size from grains to small
angular pebbles. Criteria of shape are lacking. It is 2.5 to 3.5 cm.
thick.

One sherd (fig. 2, k) from Midden C is particularly interesting,
because it appears to have been fiber- tempered. The tempering
was relatively abundant and appears to have been in clusters.
Possibly wads of grass would produce the observed conditions of the
temper molds. The paste is sandy and brown in color. The vessel
might have been elliptical. It had a flat bottom 3 cm. thick, a verti-
cal, slightly convex rim 4 cm. thick and 8.5 cm. high, and a rather
flat lip.

The sherd illustrated in figure 2, d is also from Midden C. It is
tempered with small particles of grit. The color is buff and there
are deposits of carbon, mostly on the outside of the sherd. The
vessel represented by this sherd was 9 cm. high, and probably
elliptical, with a rounded bottom 2 cm. thick, a vertical, slightly
convex rim 4.5 cm. thick, and a flattened lip.

A somewhat similar rim sherd, illustrated in figure 2, e, is made of
very sandy clay with inclusions of gravel of various sizes. The sherd
is brown with some smoke-blackening on the exterior. The rim is
vertical and straight, with a flattened lip. It is 8 cm. high and
4 cm. thick.

Another sherd (fig. 2, /) from Midden C is dark gray, coarse, and
heavily tempered with grit particles of various sizes. It is 2 cm.
thick. Criteria of shape are lacking. What appears to be the interior
surface is heavily encrusted with deposits of carbon.

Also found in Midden C was the sherd illustrated in figure 2, g.
The color is reddish except where the exterior is smoke-blackened.
The paste is composed of clay and small angular stone fragments,
with the latter predominant. The vessel of which this sherd was a

wK^w

Fig. 2. Pottery sherds from the Aleutian Islands.

Fig. 3. Pottery sherds from one vessel.

QUIMBY: ALEUTIAN POTTERY 7

part probably was elliptical or round. It was 9 cm. high, and had a
flat bottom 1.5 cm. thick, a vertical, slightly convex rim 2 cm. thick,
and a rounded lip.

In figure 2, h there is illustrated a basilar sherd with a small
portion of the adjoining side wall or rim. The paste is crumbly,
abundantly tempered with small or medium-sized grit particles
some of which are angular and others rounded. Colors are buff and
gray.

A single pottery fragment (fig. 2, i) from the D Midden suggests
a very shallow bowl with a circular outline, a curved rim, and a
flattened lip. The vessel wall is of rather uniform thickness, 3.3
cm., although it may be slightly thinner toward the bottom. The
paste is hard, well consolidated, and burned throughout. It con-
sists of abundant angular rock particles cemented together with a
little clay. Striations on the surfaces of the sherd suggest some kind
of "tooling" while the pottery was still plastic. This sherd is truly
synthetic stone and could be described paradoxically as stone tem-
pered with clay.

Two somewhat similar sherds (fig. 2, j and b) found in the G
Midden are reddish in color except where they are smoke-blackened
on the exterior. They are heavily tempered with grit. The vessel
probably was elliptical in shape. It was 7 cm. high, with a flat
bottom 2 cm. thick, a vertical, slightly convex rim from 1.5 to 2.5
cm. thick, and a flattened lip.

A sherd (fig. 2, I) from Midden A is tempered with sand. The
vessel of which it was a part must have been elliptical and rather
small, with a flattened lip. It was 8.5 cm. high, with a vertical,
slightly convex rim about 2.7 cm. thick and a bottom 1.3 cm. thick.
This sherd is buff in color and is blackened by smoke and carbon
deposits on the exterior.

The sherds illustrated in figure 3 are all from the same vessel,
found in Midden C. The vessel suggested by the sherds is a large
oval bowl with a vertical, slightly convex rim, a rounded bottom, and
a flattened lip. This bowl had an estimated maximum diameter of
about 60 cm. and a height of about 15 cm. The rim is 5 cm. thick
and the bottom 2 cm. thick.

These sherds are abundantly tempered with particles of grit and
sand. The color is buff and there are areas of smoke-blackening and
deposits of carbon, most of which are on the exterior of the vessel.
Some parts of the vessel exterior seem to be slightly corrugated in a
fashion which superficially resembles grooved-paddle stamping but

8 FIELDIANA: ANTHROPOLOGY, VOLUME 36

probably was produced by some tooling technique after the paste
had hardened.

An unusually large vessel is suggested by a buff-colored sherd
from Midden .C (fig. 4, a). I would guess that there is more grit
than clay in the paste. , The vessel indicated by these sherds was
18 cm. high, with a flat bottom 3.5 cm. thick, a vertical, slightly
convex rim 6 cm. thick, and a flattened lip. The curvature of the
sherds suggests that the vessel was round or oval, with a probable
diameter of about 75 cm.

A somewhat similar vessel is indicated by three sherds, one of
which is illustrated in figure 4, b. The sherds are dark gray, very
coarse, and heavily tempered with grit. The vessel of which they
were a part was probably elliptical with a flat bottom, a vertical,
slightly convex rim 3 cm. thick, and a rounded lip. The vessel was
12 cm. high. The sherds were found twelve feet beneath the top
of Midden C.

About one-half of a small lamp was found in Midden C. The
color is gray or smoke-black. The paste is coarse and heavily tem-
pered with very small rounded pebbles.

The pottery type suggested by the individual examples that
have been described seems to possess the following attributes. The
paste is abundantly tempered with grit and /or sand and ranges
from pottery almost indistinguishable from stone to pottery that
is easily recognizable. All of it is crude, coarse, heavy, and thick.

With the exception of one or two lamp fragments, the pottery
consists of bowls which are round, elliptical, or oval in horizontal
section. Upon the basis of cross sections these bowls can be divided
into three variants: (1) Flat-bottomed bowls with vertical, slightly
convex rims; (2) bowls with round bottoms and vertical, slightly con-
vex rims; and (3) bowls with round bottoms and outslanting, slightly
convex rims.

The function of this pottery is not known, but the carbon deposits
found on most of the vessels indicate that they were used for cook-
ing purposes. It is, of course, possible that they were lamps, but
this seems unlikely in view of the fact that the vessels are quite
unlike the stone lamps found in the middens. And at least one pot-
tery lamp found has the same shape as one of the types of stone
lamps.

TYPOLOGICAL RELATIONSHIPS OF ALEUT POTTERY

Heretofore there has been no published evidence of pottery in
the Aleutian Islands. From Atka Island, Jochelson (1925, p. 122)

Fig. 4. Pottery and stone sherds from the Aleutian Islands.

10 FIELDIANA: ANTHROPOLOGY, VOLUME 36

reports the discovery of "one unbaked clay lamp" which disinte-
grated when he touched it. In a cave at Nazan Bay on Atka Island,
Dall found a lamp carved of unbaked clay (Jochelson, loc. cit.; de
Laguna, 1934, p. 177). Neither of these lamps is really pottery.

In his excavations of an Aleut midden at Port Moller in the
Alaska Peninsula, Weyer (1930, p. 263) found a small fragment of
"red, sandy clay, which may have been baked." Although the
evidence is slight there may be pottery at Port Moller which may
resemble some of the pottery from Amaknak Island.

In her excavations at Kachemak Bay (Cook Inlet, Alaska), de
Laguna (1934, p. 68) found two small pieces of pottery "evidently
from a large vessel, roughly smoothed inside and out. The clay has
been mixed with coarse sand, and while the fragments are quite
black, especially on the inside, they do not appear to have been fired
intentionally." As well as one can judge from descriptions and
pictures, the Kachemak Bay pottery is similar to some of the pottery
from Amaknak Island.

The Aleut pottery is somewhat similar to the "unbaked pottery"
found by Hrdlicka at Our Point site on Uyak Bay, Kodiak Island
(Hrdlicka, 1944, p. 440, figs. 211, 212). This "unbaked pottery"
from the Pre-Koniag and intermediate levels seems to have the
same basic shape as the baked pottery of the Aleut.

Fired pottery was found on southwestern Kodiak Island in the
Alitak Bay region. Of this pottery Hrdlicka says (op. cit., pp. 439,
440): ". . . we came across fairly numerous fragments of baked
pottery. All the pieces were from coarse, thick, large jars, such as
occur nowhere else in Alaska. A few of the fragments showed simple
impressed border decorations." Hrdlicka seems to believe that this
pottery belongs to his Koniag level (op. cit., pp. 34, 109, 111).
These sherds were not sufficiently described for comparative pur-
poses, but the jar shape suggests a relationship with a pottery jar
from southeastern Kodiak Island which has been described by de
Laguna (1939, pp. 334-336; 1940, p. 72). In shape, at least, the
Kodiak jar is not similar to any of the pottery from the Aleutian
Islands. And as de Laguna demonstrates in two excellent papers
(1939, 1940), the Kodiak pottery is essentially similar to the Eskimo
and Indian pottery found in the area between Norton Sound and
Bristol Bay, including the Yukon and Kuskokwim rivers. All of
this pottery belongs to her "Southern group," which she believes to
have been derived in part from Asia via the Kuriles, Kamchatka, and
Aleutian chain about 1000 a.d. (1940, pp. 71, 72).

QUIMBY: ALEUTIAN POTTERY 11

AGE OF ALEUT POTTERY

The age of Aleut pottery is difficult to determine. Known depths
for the occurrence of pottery in the C Midden range from six to
twelve feet, but the rate at which Aleut middens accumulated is
not known, and consequently the position of pottery within a midden
is not a good criterion for the determination of absolute age.

If Aleut pottery should be closely related to the pottery found by
de Laguna at Kachemak Bay in Cook Inlet, it would be feasible to
use her excellent stratigraphy as a yardstick for the chronological
measure of Aleut pottery. Cook Inlet pottery is late. It appears
in Yukon Island IV which belongs to the third (and last) period of
Kachemak Bay culture (de Laguna, 1934). There are, however, so
few sherds from Kachemak Bay that it is impossible to assess the
degree of their relationship with Aleut pottery.

The Aleut pottery certainly does not seem to be related to any
of the pottery styles of the ancient northern Eskimos, although it
might be typologically related to those of modern northern Eskimos.
Should such a relationship exist, the Aleut pottery would be com-
paratively recent.

The pottery found by Hrdlicka on Kodiak Island is assigned by
him to his Koniag culture level, estimated to be from 300 to 400
years old (1944, p. 326). The Aleut pottery, however, does not seem
to be closely related to this Koniag pottery; thus, Hrdlicka's guessed
date is of only general use in determining the age of the Aleut pottery.

The resemblance of form between Aleut pottery and the "un-
baked pottery" found by Hrdlicka at Our Point on Kodiak Island
seems somewhat more suggestive of a parallel development than of a
direct connection. This "unbaked pottery" was found in the late
Pre-Koniag and Intermediate levels or periods and therefore does
not seem to be particularly early. In fact Hrdlicka's late Pre-
Koniag, Intermediate, and Koniag periods seem to equate with the
Late Aleut period in the Aleutians.

By and large, the Aleut pottery does not seem to possess strong
typological relationships with any pottery found elsewhere in Alaska.
Other investigators, however, may see typological relationships
where I see none. If Aleut pottery should turn out to be closely
related to any of the other pottery styles I have mentioned, there
would be a plausible argument for contemporaneity, and since all
of this non-Aleut pottery is relatively recent, Aleut pottery also
would be relatively recent.

12 FIELDIANA: ANTHROPOLOGY, VOLUME 36

There is a strong probability that Aleut pottery is an indigenous
development. In the first place the pottery from the C Midden
most closely resembles stone vessels from the same midden. Much
of the Aleut pottery not only resembles stone but conveys the impres-
sion that the Aleut potters regarded themselves as working in
"plastic" stone. By analogy with a stratified site (a site excavated
in such a manner that stratigraphic analysis is possible), the D Mid-
den, we know that the techniques of stone manufacture are much
earlier than those of pottery.

A preliminary analysis of the materials from Midden D shows
that it is possible to formulate an Early and a Late period of Aleut
culture. Although both have much in common there are significant
differences between them; for instance, the art of the Late period is
suggestive of Punuk and Post Punuk design in northern Alaska
whereas the art of the Early period is suggestive of Dorset design
in the eastern Arctic (Quimby, 1945a). Toggle harpoon heads of the
Early period have beds instead of slots for the attachment of stone
blades. No pottery occurs in the Early period, but at least one
large sherd comes from a Late period provenience. None of the stone
vessels occur in the Early period, although stone-working techniques
were known then and used specifically for the production of certain
styles of stone lamps. Thus all the evidence from the stratified
D Midden suggests that the stone and pottery vessels belong to the
Late period, and by analogy with the D Midden the stone and
pottery vessels from Midden C also belong to that period.

There is no available stratigraphy within the Late period which
would show the development of the pottery vessel from the stone
vessel, but typological considerations and the antiquity of stone-
working techniques lead us to assume that the stone vessels appeared
early in the Late period and that the pottery vessels were a develop-
ment from the stone vessels. In view of Hrdlicka's "unbaked"
pottery from Kodiak Island, it appears that a similar "evolution"
was in progress among the Pre-Koniags at about the same time.

The time of this development is very uncertain. By analogy
with the stratigraphy of Midden D all of the cultural content of
Midden C seems to belong to the Late period. One of the traits of
the Late period is the "sadiron-shaped" lamp of stone. Elsewhere
(Quimby, 1945b) I have described this style of lamp and have sug-
gested that it was coeval with or earlier than the "sadiron" lamps of
Kamchatka. The sadiron lamps of Kamchatka have been dated
as eleventh century (a.d.) by means of Japanese coins associated
with them. If my hypothesis is correct, then the Late period of

QUIMBY: ALEUTIAN POTTERY 13

Aleut culture dates from the eleventh century or earlier to 1741,
the time of the first Russian contact with the Aleut. And thus
Aleut pottery, apparently produced only in the Late period, belongs
to a time interval somewhere between A.D. 1000 or earlier and 1741.
To summarize briefly: The prehistoric Aleut made pottery of a
unique style that does not seem to be closely related to other known
pottery types in Alaska. Probably the pottery of the Aleut was an
outgrowth of their stone vessel industry. The development of
Aleut pottery does not seem to be much older than the beginning
of the second millennium and might be later. The pottery discussed
in this report should be described more completely and should be
analyzed by a petrographer or a similar specialist. The historical
evidence which has been presented is circumstantial and should be re-
garded as a hypothetical framework in need of considerable testing.

REFERENCES
Collins, H. B., Jr.

1937. Archaeology of St. Lawrence Island, Alaska. Smiths. Misc. Coll., vol.

96, no. 1.
1940. Outline of Eskimo prehistory. Smiths. Misc. Coll., vol. 100, pp.
533-592.

Dall, W. H.

1877. On the succession in the shell heaps of the Aleutian Islands. Contr.
N. Amer. Ethn., Dept. Interior, U. S. Geog. Geol. Surv. Rocky Mountain
Region.

Hrdlicka, Ales

1944. The anthropology of Kodiak Island. Published by Wistar Inst. Anat.
Biol., Philadelphia.

JOCHELSON, W. I.

1925. Archaeological investigations in the Aleutian Islands. Carnegie Inst.

Wash., Pub. No. 367.
1928. Archaeological investigations in Kamchatka. Carnegie Inst. Wash.,

Pub. No. 388.

Laguna, Frederica de

1934. The archaeology of Cook Inlet, Alaska. University of Pennsylvania
Press.

1939. A pottery vessel from Kodiak Island, Alaska. Amer. Antiq., vol. 4,
no. 4, pp. 334-343.

1940. Eskimo lamps and pots. Jour. Roy. Anthrop. Inst, of Great Britain
and Ireland, vol. 70, pp. 53-76.

Quimby, G. I.

1945a. Periods of prehistoric art in the Aleutian Islands. Amer. Antiq. (in

press).
1945b. The sadiron lamp of Kamchatka as a clue to the chronology of Aleut,
culture. Amer. Antiq. (in press).

Weyer, E. M.

1930. Archaeological material from the village site at Hot Springs, Port
Moller, Alaska. Amer. Mus. Nat. Hist., Anthrop. Pap., vol. 31, pp. 239-279.