Many people do not know that the Apache were just one of several mobile peoples who occupied the southern Southwest at and before
Spanish conquest. Although some researchers have suggested that the Jano, Jocome, Manso, and Suma were Apachean (e.g., Forbes
1960, and others), archaeological evidence indicates the presence of another type of group. The archaeological evidence is indicative of
a mobile group, but differs from the Apache evidence. Moreover, what has been found conforms relatively well with documentary data
provided by Spanish explorers who indicated that people with a river-focused adaptation occupied the region. The Apache were known
for their mountain focus. Thus, it should be of little surprise that the complex of archaeolgical traits found in Arizona and southern New
Mexico, that are inferred to be related to the Jano or Jocome, tend to be found near playas (seasonal lakes) and along rivers. The
following text outlines some of this evidence for a newly defined archaeological complex.

Deni J. Seymour (c) 2004-2009

see KIVA:

2009 The Canutillo Complex: Evidence of Protohistoric Mobile Occupants in the Southern Southwest. Kiva 74(4):421-446.

Early Spanish documentary sources refer to a number of mobile groups in southern Arizona and southern New
Mexico, including but not limited to the Manso, Suma, Jano, and Jocome. These groups have remained
unidentified archaeologically and their history is often conflated with that of ancestral Apachean groups.
Investigations aimed at identifying and distinguishing the suite of mobile and semi-sedentary groups who
occupied the southern Southwest has resulted in the identification of an archaeologically definable complex
that likely belongs to one or more of these historically referenced groups. Brief presentation of this suite of traits
is designed to elevate the evidence of these mobile groups from obscurity and to encourage further
investigations into the distribution and nature of this complex. Absolute dates, artifacts, and features on open
sites and in rockshelters provide the basis for definition of this complex that has remained invisible because of
its general similarity to and mixing with Archaic assemblages. This paper focuses specifically on the expression
of this complex in the southern Southwest.

Also see Jack Forbes' paper but ignore the part about the Jocome being Apache:
The Janos, Jocomes, Mansos, and Sumas Indians
Bifaces are some of the most distinctive artifacts from this period and
complex. They certainly are the most visible. Yet, in the past these biface
forms have been confused with those from the Archaic period. In fact, the
one on the right is from Ventana Cave and was listed by Sayles and
Haury as relating to the Cochise Culture or Archaic period. The one in
the middle was suggesed by MacNeish to relate to the Archaic
component at Pintada Rockshelter. They are not and even MacNeish had
to throw out numerous protohistoric dates to make this stratum and its
artifacts (including this biface) fit within his conceptualized stratigraphic
and occupational scheme for the rockshelter.

These bifaces are similar to the Harahey knives found on the plains and
hill country of Texas, including in the Trans-Pecos, and in northern
Chihuahua and Sonora.

The three forms depicted here show some of the inherent variation.
There is also another type that has yet to be described because so few
examples have been found. these seem to be most similar to the
Covington blade defined further east.
The projectile points of these groups fall within the small triangular tradition so prevalent in this region in the terminal prehistoric and
historic periods. Most often they have basal indentations and sometimes serrated margins. Minor variations in execution may be indicative
of identity, which has caused considerable confusion in the archaeological literature as archaeologists attempt to figure out which subtle
indices actually have social meaning.

For example the Soto point is highly distinctive for its "Eiffel Tower" shape, with ears flaring outward toward the base. Those points
illustrated here show this attribute rather clearly, but notice that as points are resharpened they take on a different look. These occur in
northern Chihuahua and in the southern and western portion of Texas, but are only rare in the southern Southwest. Some of these overlap
in form with the southern Arizona Huachuca point. Does this mean that there is a continuum of form that is meaningless with respect to
assignment of culture group affiliation? Does it mean that these Huachuca points were not really made by the Sobaipuri but are sometimes
found in Sobiapuri contexts because of later reoccupation of those locations, battles, and intermixing of groups?
These three points were found at the site I refer to as Santa Cruz de Gaybanipitea. This is where a documented battle took place in
1698, thereafter the Sobaipuri abandoned this valley and moved west to Sonoita Creek to avoid hostile encounter with the mobile
groups with whom they had engaged. I have always had as part of my research design for this site to find evidence of this battle in the
projectile points found there.

What is interesting about each of these is that they have individually been referred to as "Sobaipuri" by the individuals who collected
them, including some who were attempting to study the Sobaipuri. Yet, they are not. The one on the far left is side notched and
side-notched point forms are not a Sobaipuri point style. This particular specimen is consistent with forms of Apache points found
elsewhere in the southern Southwest. The middle point has excessively out-flaring ears, a trait that is consistent with Soto points
discussed above and is thought to represent the Suma or related groups. The one on the far right is fully consistent with forms found
among the Conchos, a group who occupied Chihuahua and members of which were later enlisted by the Europeans to fight against the

We know that points were lost when used and recovered by others. We know that the Apache and probably others routinely collected
points they found and reused them. We know that points were an important trade commodity. We know that ethnoarchaeological
studies have shown that points of many styles are found in a single quiver.

On the other hand, we know that many different groups occupied the area surrounding the site where these were found. Some
apparently visited this village for trade, and others came with more hostile intentions. When these points are placed beside the
documentary record and combined with evidence that many of the structures at this site show signs of burning a fairly strong case can
be constructed that this site was attacked and that these points likely represent the points tipping the arrows used in this (or an
undocumented) conflict. Archaeological data really do not get better than this, except for in the exceptional site where dead people
with arrows in them were left in the contexts in which they died.

From another site we do have an archaeological example where two individuals were found shot with 125 arrows--some of the
projectile points still lodged in their bones. Skeletal features also show evidence of lance stabbings and also dismemberment of the
hands. Possible trophy taking is also indicated by a short linear incision on the scalp that may be an indication that a small lock of
scalp hair was removed, as is ethnographically documented in this area.

This site is interesting from the standpoint of the flaked-stone assemblage found with these two individuals. Found with these two
victims was an abundant Canutillo complex assemblage, with points and bifaces. Yet even here, one of the points found was side
notched, showing again the tentative relationship between projectile point styles and ethnicity. Yet, most of the point assemblage was
of the small triangular basally indented types found among the mobile groups of the Canutillo complex. These types are commonly
confused with the Huachuca point of the Sobaipuri because there is a continuum of style and perhaps for a number of other reasons
cited above. It is also possible that when Sobaipuri and members of these other groups intermarried the traditions became intermixed
and mobile groups may have traded their points for agricultural produce provided by the Sobaipuri.

The points and bifaces found at this ambush site suggest that the attackers were Canutillo complex mobile groups, although this cannot
be stated with certainty for several reasons. First, some of the points are archetypical Huachuca points, which begs the question as to
whether the Huachuca point really is a Sobaipuri point style or whether such points are simply often found in Sobaipuri contexts and
have been confused. Yet, if "Sobaipuri" really is an amalgamation of many groups, it should not be surprising that there is a mixture of
points and a continuum of point forms.

The majority of points, however, and the bifaces, suggest the attackers were of the Canutillo complex. One additional problem is that
documentary sources occasionally mention an added dimension of humiliation wherein the victim is shot with his own weapons. Thus
we cannot be sure that the points used and the bifaces found are indicative of the attacking group, but in fact, this is the most likely
explanation and the most commonly occurring scenario in historically and ethnographically documented circumstances.

We also have examples of habitation sites with Canutillo complex materials. Small surface structures document the encampment
and suggest a mobile lifestyle. The associated flaked-stone assemblage provides further confirmation of this association.
Interestingly, this site also produced evidence of early O'odham pottery and contemporaneous pottery for which group
affiliation is not known. We know these date to the same time period because several sherds from this site were
luminescence dated. Like many mobile groups, it is possible that the Canutillo complex mobile groups did not produce their
own pottery but obtained it from their neighbors, either through raiding or trading. The earliest documentary evidence (1530s)
from the Southwest and surrounding areas indicates that raiding was taking place at this time. However, given the open
setting of this Canutillo complex sites it is suggested that in this case the pots were obtained in trade.
image of pottery, to be added
One hint as to what these people were eating is provided by their setting. They often occupied playa margins and riverside
terraces. This suggests that riparian and lacustrine resources were important--an inference that is backed up by the
documentaruy record as it relates to some of these historically documented groups. Moreover, when subjected to residue
analysis a biface (shown below) from the Sharples Site produced evidence of rabbit and fish, as well as yucca and nuts. As
expected, this biface was seemingly used as a multi-purpose tool.
Another hint about their diet is provided by a bison kill site found in southeastern Arizona. Individually killed and butchered
animals occur in a cienega or marsh setting. One in particular had a fragmentary Covington-like blade fragment associated.
Careful inspection of the flaking patterns indicates that some were hafted at the end while others were hafted in the
middle. Examples of this latter hafting technique have been found in rockshelters in Texas, as shown below.
Soto points, however, may not be as rare as we think because there is a continuum of forms between these Soto points and those
attributed to the Sobaipuri on the western edge of our study area. These Huachuca points that in the past have been attributed to the
Sobaipuri are remarcably similar to Soto points, as past researchers have pointed out. But one reason they are perceived as more
similar than they actually are is that Huachuca points have been intermixed by archaeologists with Soto points and other intermediate
forms. Archaeologists who work in southern Arizona do not broaden their studies wide enough to the east and south to understand how
widespread this continuum of forms is and how similar the so-called Sobaipuri form is to forms far soutside the Sobaipuri territory.
Moreover, Texas archaeologists similarly do not expand their studies far enough west to see that similar forms are found among the

Other reasons these are difficult to differentiate is because the groups involved were mobile and their territories encompassed huge
areas, overlapping those territories of many other groups. As a consequence, distributions of mobile group points overlap with those of
other mobile groups and with non-mobile groups. The problem of classification cannot be addressed in ways that are similar to more
sedentary groups. Culture area boundaries (and thus the distribution of material culture) are not as rigid as those of sedentary peoples.

Sobaipuri sites usually have multiple components where points of other groups were introduced at a later time. Also the Sobaipuri
interacted with these other groups, trading and intermarrying with them. Projectile points are known to have been key trade items
between groups throughout the world, and there is no reason to believe any different in this area. In fact, the documentary record
suggests a trade in projectile points. Moreover, there are documented battles between the Sobaipuri and these other groups--is there a
better way for the arrow heads of these other groups to find their way into Sobaipuri contexts?
Huachuca points, thought to be
associated with the Sobaipuri.
image of points and bifaces , to be added
These bifaces are also sometimes found on Sobaipuri sites. Usually these are found in fragmentary form and so little can
be said about them. In some instances complete or fragmentary forms have been found that have been reworked to from
some other type of tool. This is clearly the case in the examples shown below. These bifaces have been reworked to be
used as an end scraper. This is usually the preferred tool form resulting from reworking.
Prior to the current investigations only one such biface had been suggested to be associated with the Sobaipuri. This was the
biface found with the Bechtel burial. In hindsight, this burial may have been a Canutillo complex burial, like many of the other
recent burials encounted in southeastern Arizona that have been attributed to the Sobaipuri. The reason these burials, including
the Bechtel Burial, have been associated with the Sobaipuri is because this Canutillo complex had yet to be defined and it was
not known that the associated artifacts, while often found on Sobaipuri sites, are actually indicative of the mobile groups.
Front and back of a broken
reworked biface found at the
fringe of a Sobaipuri site.
Canutillo complex biface that has been
broken and reworked. Close up shows
that it was retouched to be used as a
scraper. This is from the Kuykendall Ruin,
and was found sitting on the top of the
burned roof fall layer of the pueblo,
indicating mobile groups visited this site
and lived within the ruined buildings.
Riker mount with mostly Soto points obtained from
northern Chihuahua, with a few other forms
including side-notched examples mixed in.
Examples of Soto points, all collected from
northern Chihuahua by private collectors.
Single side-notched point found with the
more than 100 small triangular points. The
missing tang was found in the collection
and the point can be reconstructed to a
complete form, minus the tip.
Hide tanning rock similar to the one
near which the biface was found.
Reworked biface found near a large
Sobaipuri site; the end has been
reworked as a scraper and
indentations retouched on the sides
allow it to be hafted like a scraper.
The broken biface has been reworked
along one margin where the original tool
broke. It is now best used as a scraper and
even includes a small indentation on one
margin that likely assisted with hafting
without cutting the sinew.
Close up of margin
Small triangular-shaped bifaces
that may have functioned as
preforms or they may have
served as points and tools
without further modification.
Expediencey is the name of the
game during this period.
This biface has been attributed to
the Sobaipuri.
to be added
Groundstone is expedient and shows only the
barest evidence of use. No attempt was made to
shape the piece. Insted, suitable rocks were
found around the site and used on a limited basis.
Anvil stones occur at the fringe of the site.
Here flaked stone is scattered around,
suggesting that these anvil stones were used
in the reduction of flaked stone material.
Excavated area with hide-working stone where biface was
recovered (blue flagging).
This is remarkably
similar to a tool found
at Ventana Cave that
also appears to have
been a reworked
broken biface.
These points are from the Sharples Site where they
date to the A.D. 1400s. They are located throughout
the occupational area at this site and mixed with
Canutillo complex artifacts and features.
An assortment of bifaces found on sites from
the Hueco Mountains and Otero Mesa (on
the east) over to Tubac (south of Tucson)
and Ventana Cave (west).
A type of arrow point referred to as Lott, from West Texas.
This Lott-looking point was also
mixed in with the others.
Comparisons to Archaic Assemblages: Bifaces

Despite past and continuing practices of lumping these later materials with those of the Archaic, there are clear quantitative and qualitative
distinctions between them. Canutillo complex bifaces and their resulting debitage may look superficially like Archaic materials simply because
they are formal tools. The discontinuation of a formal tool technology with the advent of the ceramic period has led to the erroneous grouping
all formal tools into earlier pre-ceramic stages. Yet with the rise of mobility in the late prehistoric a formal biface-based technological
organization appears again and a variety of measures allow these to be distinguished from their earlier counterparts. The goals of production
were similar (biface-oriented technologies aimed at producing durable cutting implements), but the products are quite different. When
comparing Archaic bifaces to those of the Canutillo complex one is immediately struck by basic differences in technology, visual quality, and a
number of physical attributes including the freshness of the flakes scares on the Canutillo material. These differences as discussed below
include: (1) differences in materials used (silicified limestone and fine-grained basalt for Archaic versus fine-grained silicates for Canutillo along
the same drainages), (2) differing degrees in patenation and dullness versus sharpness of edges related to age, (3) stylistic differences including
clunky, thicker, and asymmetrical Archaic bifaces versus the symmetry, high quality flaking, and thinner Canutillo ones as reflected through
visual inspection as well as in number of marginal flake scars, (4) the distinctiveness of shape between the thin ovate or leaf-shaped Canutillo
bifaces versus San Pedro bifaces, for example, that are triangular in form, with a broad or expanded base that tapers toward the opposing end,
and (5) the distinction of Canutillo complex and the Archaic San Pedro and Chiricahua materials is also apparent with regard to biface length,
width, and thickness.

When opening any drawer of the catalogued materials from the Cochise Culture type sites (Sayles' referent for Archaic period sites) at the
Arizona State Museum the clarity of the differences are immediately apparent. For example, San Pedro stage and Chiricahua stage materials
from the Fairbank and Cave Creek type sites, respectively are heavily patenated, with rounded edges that have been dulled with age (Figure 6).
These represent the latest materials from the Cochise culture sequence; earlier Archaic materials are even more weathered (but few bifaces
are included in the collections). In comparison, protohistoric tools are fresh looking, often with a bright luster to the surface and with relative
sharp edges. Even on sites with both prehistoric (Archaic or ceramic period) and Canutillo complex materials, the protohistoric tools and
debitage consistently show substantially less weathering than their earlier counterparts, whether in buried or surface contexts. This is not
surprising given the late dates that are produced on sites containing these unweathered artifacts from across the region.

Materials Used
Material types used were also distinctive. Archaic-age artifacts from the San Pedro Valley and much of southeastern Arizona are often made of
silicified limestone and fine-grained basalt whereas protohistoric Canutillo complex bifaces are made of fine-grained cherts, chalcedonies, and
fine-grained rhyolites. Through time these materials seem to be locally procured, including the red, orange-tan, and gray cherts that are found
in neighboring mountain ranges along the San Pedro, in tributary canyons, and in lag gravels. The selection of distinctive material types in
different eras likely relates in part to the quality of the desired finished product; large pieces of finer-grained materials were needed to achieve
the thin final protohistoric specimens that were produced through fine bifacial flaking.

Style and Symmetry
Stylistically the symmetry of the protohistoric Canutillo complex bifaces is immediately apparent, whereas the Archaic bifaces are often clunky
and asymmetrical. The difference in flaking quality and final product probably relates to the use of Canutillo complex bifaces as knives and
perhaps as flake cores, whereas the thicker Chiricahua stage and San Pedro stage Archaic ones probably functioned as cores for the production
of flakes (see Sayles and Antevs 1941: Plate Xd; Haury 1975:272, Figure 56d, 276, Figure 57d). The thinner San Pedro stage bifaces (that were not
projectile points) were probably used as cutting implements (Haury 1975:269, Figure 54e,f; also see Huckell 1984:Figure 5.23k,l,o; Huckell
1995:55, Figure 4.2d,e), but these are not as finely flaked, nor as thin, or as symmetrical as the later forms (Haury 1975:265, Figure 52f).
Moreover, Canutillo bifaces are alternately flaked on opposite sides of the tool so that the edge is wavy and produces a distinctive and durable
cutting edge.

Flaking Quality
This difference in flaking quality is quantifiable in the amount of flaking on tool surfaces, which differs substantially between Archaic and
Canutillo protohistoric bifaces. This is apparent with simple visual inspection but can be quantified with a measure of number of marginal flake
scars per a 10 mm linear area along a randomly selected portion of the margin. Figure 8 illustrates the differences in quality of manufacture
and investment in flaking that characterize this tool form through time. San Pedro bifaces-those from the Archaic that are most similar to the
Canutillo complex-average around 1 to 2 flake scars per 10 cm area whereas Canutillo bifaces have three to five. Finer flaking likely relates to
the desire to maintain flaking control so as to produce the desired edge on and thinness to the tool.

The distinctive ovate or leaf-shaped form of Canutillo bifaces is one of the most telling attributes, particularly when combined with specimen
thinness. A subset of San Pedro bifaces, those that probably functioned as knives, is most similar to Canutillo bifaces with regard to their thin
profile. But as shown in Figure 9, these San Pedro bifaces are triangular in form, with a broad or expanded base that tapers toward the
opposing end. Plus, while Chiricahua bifaces trend toward an ovate or leaf-shaped form, these are substantially thicker than their protohistoric
counterparts, with deep multidirectional negative flake scars that are consistent with these forms being used as cores or heavy implements.
Many of the thick San Pedro bifaces are also ovate but, like Chiricahua bifaces, they lack symmetry in flake scar orientation and probably
functioned as cores or more durable forms of tools. The thin Archaic and Protohistoric bifaces that possess attributes indicative of their use as
knives (e.g., thinness of margins consistent with cutting and acute edge angles), are distinctive in outline form as compared to the Protohistoric
ones. Although most Archaic knives are broken, the tapering nature of their margins is indicated on the fragmentary specimens and is
confirmed in the more complete forms.

Canutillo bifaces tend to fall into at least two shape categories suggesting that there may be grounds for further subdivision. These differences
probably relate to how they were hafted. One variety has an ovate outline (with its greatest width near the midsection) and two pointed ends,
and may have been hafted as shown in Figure 10a. The second is pointed on one end and blunt or flattened on the other suggesting hafting
from the end (Figure 10b). Evidence for these differences in hafting is provided in the negative flake scars on the surface of these specimens. An
often-visible feature of these bifaces is a slight indentation on one or both margins, where the edge was blunted so that the sinew would not be
severed. In addition, flake removal on the blunt end of some specimens indicates preparation for hafting. More obviously, the face of the double
pointed bifaces is often thicker near the center of the tool where flakes seem to have been intentionally terminated (hinged out) in the center to
form a thicker section to hold the hafting in place. This placement is consistent with the indentations on the margins. In some instances, instead
of being raised one or more large flake scars will decrease the thickness of the surface, leaving an adjacent raised surface, which accomplishes
the same result (Figure 11). This effect can be quite subtle and were it not for photographs showing hafted examples (Figure 12) this attribute
might go unnoticed.

It is likely that these bifaces were used as knives, judging from the hafted examples, the vaguely sinuous margins, and the acute edge angles.
This inference that is consistent with the historic record that notes the use of knives among mobile groups encountered in southern New Mexico
and Texas along the Rio Grande (Ayer 1965:14; Hodge and Lewis 1990), as was discussed briefly above.

As Figure 13a-c shows, basic metric differences between Canutillo and Arizona Archaic (the original Cochise culture complex type site)
materials demonstrate they differ in more than visual qualities. The sample size is still small because the protohistoric complex has only
recently been defined. Also, regrettably, these fine tools are the type of items that have been removed from sites by collectors because they are
pretty, valued, and visible. Although these types of tools have been found in sound archaeological contexts, many specimens are known from
collector's riker mounts or donated unprovenienced finds in museum drawers. Some, like those from Ventana Cave and Pintada Rockselter, are
included in mixed assemblages that have not yet been culled. Regardless of these difficulties, the metrics demonstrate discernable differences
between these protohistoric bifaces and those attributable to the Archaic period. Thus, for those not convinced of the difference between
Canutillo and the Archaic by the aesthetics of the tools, the measurements provided in the accompanying figures supplement the visual
qualities in characterizing the differences.

The distinction of protohistoric Canutillo and the Archaic San Pedro and Chiricahua materials is apparent in these figures. With regard to biface
length, width, and thickness tangible differences can be seen graphically (Figure 13a-c). While Canutillo and Chiricahua bifaces are most
similar with regard to length and width, they are most different in thickness. San Pedro and Canutillo bifaces are most similar in thickness, as is
consistent with their inferred use as cutting implements, but they diverge widely in length and width.

Figure 14 illustrates the relationship between biface length and thickness. From the constructed index (length divided by thickness) it is possible
to see that Canutillo complex bifaces are at the same time longer and thinner than their Archaic counterparts and in this relationship there is
little difference between San Pedro and Chiricahua stage materials. Canutillo complex bifaces are so distinctive because they are relatively
thin for their length as compared to San Pedro and Chiricahua stage materials.

By a variety of measures and considerations the Canutillo complex materials are distinctive from the Archaic assemblages that are found in the
same geographic areas, often in similar environmental and physiographic settings, and many times on the same multicomponent sites. These
bifaces are but one of the many tool types found on these Canutillo complex sites but they serve as an example of the uniqueness of this
protohistoric assemblage from the Archaic assemblage with which they have been confused.
These figures and the associated text were originally part of the Canutillo complex Kiva article
(see reference at end) but made the article too long so they were edited out.
Figures 6 and 7. These figures compare Canutillo complex bifaces on the far left to Chiricahua stage Archaic bifaces
on the right, and San Pedro Stage Bifaced in the separate figure to the far right. The degree of weathering is being
shown. The difference in weathering between the periods (Archaic versus Protohistoric) is obvious as is the
difference in the material types used. The shapes of the points and the crafting of the points differ substantially too.
Figure 8. Flaking quality was examined by measuring in
number of flake scars along the margins of tools
between Canutillo complex tools (one on the left) and
Cochise culture ones (three on the right). Something is
lost in the translation to low resolution Jpeg, making
these a bit blurry, but the point should be clear.
Figure 13 a-c. shows thebasic metric differences between Canutillo and Arizona Archaic.
See Metrics discussion above.
Figure 14 illustrates the relationship between
biface length and thickness, as discussed above.
Flake scars intentionally hinge out in the middle to facilitate hafting.