Redman, Kimberly L. and Noel D. Justice (Glenn A. Black Laboratory of Archaeology,
Indiana University, Bloomington)
MCKINLEY SITE: A LATE ARCHAIC BURNED ROCK MIDDEN IN CENTRAL INDIANA, CURRENT COLLECTIONS
PROCESSING AND RESEARCH
Initial excavations at the McKinley site in the 1950's demonstrated a major occupation
dating to the Late Archaic period with traces of earlier, as well as later, occupations.
Before proceeding, we would like to point out that the site was initially discovered
and reported by workmen connected with the building of highway 37 in 1933 and
its improvement in 1950. It was their watchful eyes and curiosity that eventually
gained the attention of Glenn A. Black and Jack Householder. Again in the 1970's,
concerned non-professionals took responsibility and initiated rescue excavations
with knowledge that road relocation might seriously endanger the site. It is due
to the efforts of the Indianapolis Amateur Archaeological Association that the
material from the latest excavations are now available for study. Members of that
team spent countless hours in the field and at the Glenn A. Black Laboratory to
process flotation samples and to re-bag and sort the remains.
This report focuses on the 1970's excavations at McKinley site. We will discuss
the processing and cataloging of the material remains and some interesting aspects
of the collection. We will also attempt to characterize certain spatial relationships
based on the information from the combined excavations at the site.
When the entire collection reached the laboratory it required approximately 100
cubic feet of storage space. There were some attempts to reduce this amount in
the course of graduate research by John Richardson and Ruth Brinker, but the collection
remained largely unprocessed and uncataloged until four years ago.
When the senior author became involved with the McKinley site, there were approximately
450 catalog cards which had been completed mainly on material from features, and
a number of bags of fire cracked rocks had been sorted. The first major task was
to organize the remaining collection. This involved separating bags of material
by original unit and level provenience. The deteriorated condition of some original
paper bags required rebagging into plastic containers. Fortunately, only a small
amount of information had been lost. Upon investigation of the contents of the
bag, it became immediately apparent that the collection was primarily composed
of fire-cracked rocks with much smaller amounts of cultural items such as chipped
and ground stone tool fragments, burned and unburned bone, charred wood and nut
shells, ceramics, and mollusk shell.
It was necessary to reduce the vast quantities of cultural debris into meaningful
units in order to ensure total recovery of important cultural items and to reduce
the amount of material. After several preliminary trials, a sorting strategy was
established to process the collection in a structured manner, which was then applied
to each bag of material in preparation for cataloging and analysis. Some materials
required washing to ensure proper identification of artifact type and raw material.
Items of a questionable nature were compared directly to diagnostic artifacts
from other Archaic and Woodland occupations to establish their identity.
At the McKinley site, prehistoric Indian groups used cooking methods that involved
the collection of glacial rocks for stone-boiling, warming stones for heat transfer,
and baking stones for earth ovens. The process of repeated heating and cooling
fractured the rock, forming what we classify today as fire-cracked rock or FCR.
By its nature, fire- cracked rock might appear to be limited in analytical value.
Historically, it was common practice for archaeologists to discard vast quantities
of this material without any particular record, as was the case with the early
excavation of McKinley. Due to the large quantity of FCR in the collection, we
decided that it was important to analyze this material as carefully as possible
before discarding. In order to insure 100% recovery of prehistoric tools and other
important cultural items, the contents of all bags were thoroughly separated.
All FCR was inspected for any evidence of prehistoric use. A total of 1.3 tons
of unmodified FCR was separated into categories of sandstone, limestone, and hardstone
or granite. Unmodified granitic FCR dominated the collection. Hardstone represented
91.4% of the rock in the unmodified FCR category, while sandstone and limestone
each represented only 1% (Figure 10).
The portion of the FCR that was culturally modified is represented by fragments
of hammerstones, possible ground stone tools, battered stones and an axe or maul
fragment. Items placed in the modified FCR category are those that were tools
or showed evidence of use before or after being thermally fractured. In other
words, what would appear to be residual debris resulting only from cooking activities
actually contained materials representing any number of activities. Identifying
ground and battered tool fragments out of the FCR became a main focus of our sorting.
Those items which were identified as having distinct or possible use-wear were
saved, while unmodified FCR was weighed and discarded. Weight and provenience
information for each of the categories of unmodified FCR was documented and included
in the collection catalog for the site. Marl, a soft chalky glacial rock formed
from pulverized and redeposited limestone, was found throughout the separation
and cataloging process. All types of FCR and marl were common throughout the entire
1977-79 collection.
The items that represented the categories with the next highest frequency in the
McKinley site midden were chert, bone, and fired-clay. Examples of chert materials
found at McKinley site are glacial chert pebbles, heat-fractured chert, chert
flakes, and chipped stone tools. Glacial chert and heat-fractured chert were present
in almost every unit and level, but not as often in features. Chert flakes were
present within excavation units and sometimes in features, but in much smaller
quantities. Chipped stone tools were present atthe site in the form of projectile
points, drills, cores, and bifaces. Flaked stone tools were most often found in
the plowzone and first excavation level, although some of the diagnostic projectile
points and drills were found in features. Animal bones were present in most units,
levels, and some features. The most common animal represented was the white-tailed
deer. Other animals that are present in the bone collection are raccoon, turtle,
rabbit, and beaver, although a thorough analysis of bone from the 1977-79 excavations
has not been completed. The faunal analysis completed by Little and Adams (Little
1970) for the early excavations at McKinley site include all of the above animals
as well as ground hog, squirrel, opossum, muskrat, domestic dog, various birds,
and rodents. Very few pieces of identifiable pottery were recovered at the site.
One fragment of a grooved axe was found in a bag with "unmodified" FCR from the
McKinley site. This fragment was found in close proximity to the area of the excavations
from the 1950's. Two sides of the item have characteristics comparable to fire-cracked
rock. The other side, however, shows the distinct groove and blade portion of
an axe (Figure 11 a). The full-grooved
axe appeared in the prehistoric record at about 5000 B.C. and the three- quarter-grooved
axed appeared to be an improved invention beginning about 3500 B.C. Both types
of axes occurred through the Late Archaic period.
Although no atlatl weights were found in the 1970's excavations, there were pieces
of fire-cracked rock with petrologic characteristics that compare well with raw
materials that were prehistorically selected for atlatl weight production. Several
fragments of schist were identified and appear to be the same kind of material
as that used in the manufacture of an atlatl weight fragment from the 1953-55
McKinley excavation (Figure -11 b). This
schist material is also very close or the same as that used for atlatl weights
from the Elrod and Clark's Point sites near the Falls of the Ohio in Clark County,
Indiana. While the origin of this material either from glacial till or some unknown
quarry is unknown, we believe that the presence of this material at Late Archaic
sites in southcentral Indiana and within the Ohio River Valley is not coincidental.
Atlatl weight production was undoubtedly highly labor- intensive compared to other
tool making activities, and raw materials were selected more for aesthetic or
other qualities than for workability. This suggests more than a superficial connection
between McKinley and Late Archaic sites to the Southeast (see discussion by Justice,
this publication). Here is an example of how important it is to carefully inspect
and identify samples of fire-cracked rock.
A range of projectile points and other diagnostic tools were recovered during
the 1977-79 excavations at the McKinley site. A few examples of Late Woodland
Madison triangular points were identified. An example of a Madison point drill
was found (Figure 11 c). There is a predominance
of Late Archaic Stemmed cluster points, most of which are of the McWhinney Heavy
Stemmed type (Figure 11 e). There are
also drills, and other fragmented and resharpened Archaic tools. Two Matanzas
points, one Raddatz Side Notched and several Early Archaic Kirk Corner Notched
cluster points were also found. One projectile point of particular interest is
a specimen of an off-white, unidentified chert that can be classed as a Palmer
Corner Notched. This type is a member of the Kirk Corner Notched cluster (Justice
1987) and exhibits heavy grinding of the basal edge (see Figure
11 f). This specimen was found in Feature 47 at the site. This feature requires
radiocarbon dating to test what may be evidence for a discrete Early Archaic component
at McKinley site. In the paper by Justice (this publication) it was noted that
all evidence of Early and Middle Archaic occupations from the 1950's excavations
was from disturbed contexts in the midden.
Bone tool fragments such as awls, needles, and cut and ground antler were sorted
from the debris in the 1977-79 collections. Comparisons between specimens from
the 1950's and the 1970's excavations were made. Both excavations produced splinters
of deer bone that were worked, shaped, polished and drilled; most show heavy use.
Some of the specimens can be classified as needles and probably relate to sewing
or weaving activities. A specimen which was broken through the weak point in the
eye was found in 1977-79 excavations (see Figure
11g). Worked deer antler tines were recovered throughout excavations at the
McKinley site (Figure 11 i) . These had
many functions before fracture and discard (e.g. projectile points, flint knapping
tools, wedges, tool handles, etc.) . One item from the more recent excavations
shows a striking similarity to the broken atlatl hook that was recovered from
the early excavations at the site (see Justice, this publication). While the 1970's
specimen has been subjected to heavy weathering, it has many of the same characteristics
seen in better preserved examples. Fragments of bone awls were also found
(Figure 11 h).
The identification of ceramics in a collection from a predominantly Late Archaic
site was at first problematical. It was thought that potential disturbances might
have obscured the cultural integrity of the site. Instead, we positively identified
a late prehistoric component at the site. Many sherds were quite small and eroded,
thus it was sometimes difficult to segregate actual ceramics from incidental pieces
of burned clay. There were a few large, thick examples of burned clay or ceramic
material that have a coarse temper, which leads us to question their cultural
identity. While there are only three examples, certain sites in Missouri (Reid
1984) and Tennessee (Peterson 1973) have produced meager evidence of crude ceramic
technologies that appear in otherwise Late Archaic contexts. Farther afield at
a site in Kansas, there are rudimentary ceramic figurines and fired clay at a
site dated to around 3500 B.C. (Witty 1982). A few of the McKinley specimens could
represent an undefined Early Woodland ceramic, like Marion Thick (Helmen 1951),
or these too could simply be incidental, burned clay. Certain projectile points
found at McKinley add to the suggestion of an ephemeral Early Woodland occupation;
therefore, we should not rule out the possibility of early ceramic technology
at McKinley. Baring this cautionary note, it is apparent that nearly all identifiable
ceramics (see Figure 11 d) relate to
the late prehistoric, Oliver phase defined by Helmen (1951) (see also, Dorwin
1971; McCullough 1991). The 1977-79 excavations confirmed that this component
extends over the entirety of the McKinley site.
Some analysis has been completed on the debris of the McKinley site excavation
units. We have found that significant variation in the amounts of FCR occurs between
units indicating some segregation of activity areas at the site involving the
use of glacial rock in the process of cooking and roasting food. All combined
information suggests repeated short-term occupations of the McKinley site. Basically
all of the descriptions of the midden, features and contents from the earlier
excavations apply to the 1977-79 collections. There were features filled with
FCR, and there was an abundance of dark organic soil with charred wood and nut
shells along with some mussel and snail shells.
In summary, the 1977-79 collection is now finally cataloged, 15 years after its
recovery, and is being entered into the Glenn A. Black Laboratory on-line database.
A computer generated map of the site is also being prepared. McKinley site is
culturally diverse, but it has a major Late Archaic component with social and
cultural affinities to sites in the Falls of the Ohio Region in Indiana and Kentucky
and to sites associated with the "Central Ohio Valley Archaic" study unit defined
by Vickery (1980) for southwestern Ohio. There is a less intensive late prehistoric
occupation, and there are indications of discrete features which may date to the
Early Archaic period. In all, the site and collections have tremendous analytical
potential for local and regional archaeological problems. In addition, there are
significant parts of the site that remain preserved for the future.
References Cited
Dorwin, John T.
1971 The Bowen Site: An Archaeological Study of Culture
Process in the Late Prehistory of Central Indiana.
Indiana Historical Society Prehistory Research Series
Vol. 4(4).
Helmen, Vernon R.
1951 The Cultural Affiliation and Relationship of the Oliver
Farm Site, Marion County, Indiana. Unpublished Master's
thesis, Department of Anthropology, Indiana University,
Bloomington.
Justice, Noel D.
1987 Stone Age Spear and Arrow Points of the Midcontinental
and Eastern United States. Indiana University Press,
Bloomington.
1993 Cultural and Historical Perspectives of the McKinley Site.
This publication.
Little, Robert M.
1970 The McKinley Site. Unpublished Master's thesis,
Department of Anthropology, Indiana University, Bloomington.
McCullough, Robert G.
1991 A Reanalysis of Ceramics from the Bowen Site: Implications
for Defining the Oliver Phase of Central Indiana.
Unpublished Master's thesis, Department of Anthropology,
Ball State University, Muncie, Indiana.
Peterson, Drexel A.
1973 The Spring Creek Site, Perry County, Tennessee: Report of
the 1972-1973 Excavations. Memphis State University
Anthropological Research Center Occasional Papers No. 7.
Reid, Kenneth C.
1984 Nebo Hill and Late Archaic Prehistory on the Southern
Prairie Peninsula University of Kansas Publications in
Anthropology No. 15.
Vickery, Kent D.
1980 Preliminary Definition of Archaic "Study Units" in
Southwestern Ohio. Prepared for the State Archaeological
Preservation Plan Meeting, Columbus, Ohio.
Witty, Thomas A., Jr.
1982 The Slough Creek, Two Dog, and William Young Sites,
Council Grove Lake, Kansas. Kansas State Historical
Society Anthropological Series No. 10.
Key to Figure 11
a. Grooved axe fragment, granite. N 940 E 120, depth 0-.6'.
GBL 6110/3489.
b. Prizmoidal atlatl weight fragment, schist. 7 1/2 R 3,
depth 0 .4'. 1953-55 excavations, GBL 872/161.
c. Late Woodland, Madison drill form; Liston Creek chert?
Surface, east trench. GBL 6110/621-1.
d. Rim sherd; finger-impressed, grit temper. N 890 E 5,
depth 6 12", level 2. GBL 6110/2446.
e. McWhinney Heavy Stemmed point base; unidentified glacial?
chert. N 890 E 5, depth 6-12", level 2. GBL 6110/2579.
f. Palmer Corner Notched; Silurian chert? Feature 47, 11"
below level. GBL 6110/4352.
g. Needle or bodkin; carved animal long bone, polished and
damaged from use. N 860 E 25, feature 22. GBL 6110/4224.
h. Needle or awl tip; split animal long bone, polished and
worn from use. N 855 E 25, feature 27, 18-24".
GBL 6110/4299.
i. Antler tine; snapped, carved and shaped. N860 E 25,
feature 22. GBL 6110/65.
[return to 1993 abstracts menu]
[continue to next]