Ocmulgee Old Fields

After the Spanish exploration of Georgia in 1540, about 150 years

elapsed before the Ocmulgee tribe of the Creek Nation settled at a place

which we can now identify with reasonable certainty. This site in later

years was known as Ocmulgee Old Fields, for the evidence of ancient

cultivation can often be detected long after the signs of dwellings

themselves have disappeared. Needless to say, this was the last Indian

village of any importance to occupy the area now included in Ocmulgee

National Monument.

The recognition of this village site was partly brought about by the

intensive study of an interesting feature of Colonial construction

disclosed early in the excavations. This consisted of a ditch about 1

foot wide by 2 in depth which outlined a curiously shaped area on the

Macon Plateau some 200 yards north of the Great Temple Mound. Presumably

the footing ditch for a palisade, it enclosed a space shaped like the

gable end of a house with very low walls and a steep roof. The base

side, facing northwest, was about 140 feet long and was interrupted at

two points, suggesting a large central entrance gate with a smaller

postern 18 feet to the left. Surrounding the enclosure on all but its

long base side was a broad, shallow ditch which may have served as a

moat. It might, though, have been used instead to improve the drainage

of the stockade; for excavation showed that this lay close to old

springs which had once issued from the adjacent high ground. Finally,

the remains of a wide beaten trail from the northeast, worn a foot or

two into the old land surface, were found to terminate before the

entrance. This path had been traced at intervals across the plateau for

about half a mile, and was picked up again beyond the enclosed area

leading off down the hill toward the river.

Inside the stockade, rectangular blackened areas in the soil indicated

what appeared to be the decayed remains of several log buildings, while

mixed with the usual debris of an Indian village site were numerous

articles of European manufacture. Both here and at other points, chiefly

concentrated on the southwest corner of the Macon Plateau, excavation

revealed iron axes, clay pipes, trade beads, brass and copper bells,

knives, swords, bullets, flints, pistols, and muskets. All indications

pointed to a large and thriving Indian community situated generally at

the western edge of the old Master Farmer village site, and plentifully

supplied with English trade materials. The fact that a small fortified

structure existed in the midst of this community at once suggests the

very trading post from which these goods were obtained.

Returning, now, to the Early Creeks, we left them sharing in the

development of a distinctive material culture which characterized, with

minor differences, a large portion of the Southeast. When we encounter

their descendants on the site of Ocmulgee Old Fields, however, we find a

mixture of old elements and new; and it is often difficult to say what

part of the changes we observe was due to European contacts and what a

normal continuation of the development which had gone before.

[Illustration: Excavation of Trading Post stockade. Darkened soil,

indicating position of log wall footing, emphasized to show gates (right

half of long wall, top of picture).]

The remains of this period were found thickly scattered about the

Funeral Mound, between the latter and the Great Temple Mound, and about

the area of the Trading Post itself, as we shall term the fortified

enclosure, anticipating further discussion. They consisted of burials,

pits filled with refuse, oval patterns of post molds indicating small

house sites (although these were found only in and around the stockaded

enclosure), and refuse of all sorts scattered about on the general level

of the occupation. The burials were made both within the main village

area and about the Funeral Mound, where the signs of habitation may have

been destroyed by plowing. Usually the dead were buried in a flexed

position shortly after death, and were not subsequently moved. This was

true, also, of the earlier Lamar occupation, but in marked contrast to

the Master Farmer custom of secondary burial, or the reinterment of

bones already once buried or otherwise put away.

[Illustration: Beads, gun cocks, flints, lead shot, knives, pipes, brass

bells, and other trade goods show contact between Creeks and English.]

The pottery of these historic Creeks shows that they had finally given

up the ancient habit of complicated stamping. This seems all the more

curious when we reflect that their neighbors and enemies, the Cherokee,

retained this idea, as previously mentioned, until they finally gave up

pottery entirely. In place of it, the Creeks roughened many of their

pots by brushing or stippling the surface, probably with a handful of

small twigs or pine needles. The carinated bowl form was retained,

however, along with deeper jars and other more common shapes of former

times; and on its shoulder appeared a weak, thin incising, often hardly

more than a series of crude scratches. Still, the interlocking scroll

seems to have continued as one of the basic design ideas; but it was

crudely executed, as were the hatched elements of parallel lines which

were no longer carefully integrated with the remainder of the design.

One gains the feeling that the potter was striving for the same general

effect, but was no longer interested in achieving the precision of

pattern and boldness of line on which that effect originally depended.

The lower parts of these bowls are now smooth, and many vessels are made

without either decoration or roughening.

[Illustration: Creek pottery continued some of the more characteristic

older shapes, but the decoration was only a rough imitation of earlier

designs.]

Other artifacts suggest the increasing reliance on European goods

supplied by the traders, which we know had already begun to destroy the

rude but effective and self-sufficient culture of the Indians. A highly

prized musket cost a man 25 deer skins; but once he had it, with bullets

at 40 to the skin and powder 1 skin to the pound, he could kill more

deer and would have little need to make arrow heads of flint. Another 4

skins would purchase an ax, 4 more a hoe; and again he had better, more

lasting tools without the work of making them and constantly replacing

them. Small wonder that stone tools and weapons become less frequent,

and that flint chipping itself, within a few generations, had become a

lost art.

While the Trading Post site has not yet been studied in detail, one gets

the impression that stone tools are actually less numerous. Projectile

points are mostly of small size, often very narrow triangles less than

an inch in length. European materials like gun flints and bottle glass

are used for scrapers. Glass trade heads are mixed with those made from

the central core of the big marine whelk, commonly called “conch.” Sheet

copper is used for decorative cuff-like arm bands; frequently it is

rolled into small cone-shaped janglers which were probably sewn to

clothing in clusters to replace the old deer hoof rattles.

The Indian trade was the most effective weapon of the English in their

contest with Spain and France for control of the southern frontier.

Indirect evidence points to the establishment of a trading post in this

vicinity about 1690 by the Charleston traders. Apparently lured by the

prospect of English goods, a number of the principal Creek villages had

moved about this time from the Chattahoochee, close to the Spanish

settlements in west Florida, to the Altamaha and its western fork, the

Ocmulgee. No direct reference to the position of the Ocmulgee town

during this period has yet been found, though in 1675 and again after

1717 it was reported on the Chattahoochee. Nevertheless, the Ocmulgee

are listed among Creek towns in this vicinity, and the river appears to

be called by this name as early as 1704-5.

Early maps show the site of Macon to be occupied by the Hitchiti, a

tribe of the Creek nation whose speech was older in Georgia than the

Muskogean of the true Creeks. The Ocmulgee also spoke Hitchiti, and a

Creek legend, recorded much later, states that the Hitchiti were the

“first to settle at the site of Okmulgi town, an ancient capital of the

Creek confederacy.” Legend also named the Ocmulgee fields as the first

town where the Creeks “sat down ... or established themselves, after

their emigration from the west.” This identification, made at a time

when the Old Fields were still in Creek territory, leaves little doubt

that the Ocmulgee tribe itself had once lived here. The further evidence

of a stockade erected here in a pattern then common to Colonial

fortifications in the Southeast, plus the quantities of trade material

in the village area, make it reasonably certain that this was another

center of the little recorded trading enterprise so important to

England’s success in the race for colonial territory.

While it cannot be considered evidence in identifying the site, two

additional bits of historic detail add interest to the part it may have

played in the important events of this period. In December 1703, Col.

James Moore set out upon a mission for the Carolina Assembly to destroy

the Apalachee Indian settlements in west Florida. These Indians lived in

agricultural communities, close to the missions where they received

their religious indoctrination from the Spanish. They supplied important

provisions for both St. Augustine and Havana; and their area served as a

base for Spanish efforts to win over the Creek Indians to the north and

so enlarge their dominion. Moore took with him 50 volunteers from

Charleston, and gathering 1,000 Creek warriors on the Ocmulgee, set off

southward for Florida. The raid proved highly successful and, with

others like it over the following 2 years, dealt a devastating blow to

Spanish colonial aims. By 1706, most of the Indian population of the

area had been killed, driven away, or captured. Carolina was secure from

inland attack, and Spain’s efforts to enlarge her hold in the Southeast

were at an end for all time.

Returning now to Ocmulgee, we note that as late as 1828 a map of this

region shows “Moore’s Trail” running down the west bank of the river

from a point about 2 miles below Macon. It is not hard, then, to imagine

the former governor of the colony setting off on that bold adventure. A

shouting horde of excited Creek warriors assembles near the trader’s

store. Moore watches as they fall in behind his sturdy band of

Englishmen, and the line files past the high walls of the stockade.

Following the Lower Patch down the hill, they march in the very shadow

of that imposing relic of former days, the Great Temple Mound. Then a

little distance down the river they reach the fording place; and

crossing it, are lost to sight as they enter the woods and take up the

trail to the south.

This episode, however, was merely the beginning of the Indian’s unhappy

involvement in the rivalries of European nations and of the destruction

of his own culture through his very eagerness to obtain the wonderful

products of those nations. More and more the trader’s goods were to

become a necessity to him rather than a luxury. His life shifted from

that of a village farmer to that of a hunter who left his village for

months at a time in search of the deer skins on which the new barter

economy was based. The women folk, of course, stayed home and tended the

fields; but the old ways were steadily breaking down. Moundbuilding had

been given up even before the coming of the English. With the barter

economy, the religious festivals connected with the farming calendar

were also abandoned during the prolonged hunting season. Only the great

summer harvest festival, the “busk,” or “poskita,” remained as the

central element of Creek religion. Finally, after the deer had been

largely hunted out and the market for skins had almost dried up, the

Indian became at last a log cabin farmer, exploited, but otherwise much

like any other resident of the frontier.

[Illustration: Creek warriors join the Carolina volunteers at Ocmulgee

for the start of Moore’s raid, 1703.]

Two scenes in the story of the Indians at Ocmulgee remain to be

described. All along the Atlantic seaboard the red man awakened at last

to his peril: the land hunger of the foreigners was insatiable, and in

it lay a threat to his very existence. If he could only have brought

himself to forget old rivalries and have joined his ancient enemies in

the common cause, perhaps they could have driven out the intruder before

it became too late. Sooner or later bloody uprisings took place in most

sections of the country, but the end was always the same. The old habits

were too strong; cooperation could not replace hostility overnight; the

Indian could not make the needed sacrifice though his life was at stake.

The Southeastern uprising was called the Yamassee War in which many of

the shattered tribes of Georgia and Alabama took part. There can be

little doubt that the Creeks, under the able command of their leader,

Brim, were the principal actors. Under his guidance, they had at first

helped the English against the better entrenched Spaniard, but now it

was the Englishman himself who posed the chief threat and who must be

driven out at any cost. The scheme was well planned—and came within a

hair’s breadth of success. It depended on winning over the Cherokee, who

from the first had befriended the Charleston colonists; and to do this

Brim took the unprecedented step of sending emissaries to his old enemy.

If they had agreed to forget old hatreds, the Indians could easily have

massed the strength to drive the colonists into the sea. Instead, the

Cherokee council voted to stand by their old friends; and the

announcement of their decision was the slaughter of the Creek

emissaries.

Nevertheless, the others decided to carry on without them. Their first

act in 1715 was to kill off the traders scattered about the Creek Nation

and to attack outlying settlements. Here, we can be sure, the trader to

Ocmulgee lost his life, unless by good fortune he happened to have gone

to Charleston to lay in supplies. In any case, the existence of the

store must have terminated. Little more was accomplished, however; the

Creek design had failed, and in 1717 the war came to an end. Ocmulgee

and the other towns along the river were then too close to the English

settlements at Augusta, and the Indians moved their villages back to the

Chattahoochee.

About 1773 we have a vivid description of the mounds and of extensive

old fields along the river, but there is no mention of Indians living

anywhere near the site. It is then, however, that we first learn of the

high regard of the Creeks for this spot; for here it was, according to

tradition, that the confederacy was first established. In 1805 the

Creeks ceded to the United States most of the lands bordering the

Ocmulgee River on the cast; but in this treaty they specifically

reserved for themselves about 15 square miles encompassing the site of

Ocmulgee Old Fields, though allowing the Government the right to erect a

military post or trading house thereon. In 1806, Fort Hawkins was built

a short way to the north of the Plateau on high ground commanding the

river. It was designed as a frontier outpost and factory, or trading

house; and it served this end until 1817, when it was moved west to Fort

Mitchell in Alabama Territory to keep up with the movement of the

frontier. Once more the Indians gathered here in 1819 to receive the

annual payment for their lands east of the Ocmulgee; but the city of

Macon was founded only 4 years later, and in 1828 Ocmulgee Old Fields

was sold to the public.

Ocmulgee National Monument, Georgia · 第 11 / 18 题