Lithic raw material in the vicinity of the Chivay source

Variability in material type throughout the survey area make clear the basic structure of lithic procurement in the area of the Chivay source.

Figure 6-1. Artifactual lithic material types in theUpper Colca Project study region.

	Obsidian		Volcanics		Chalcedony		Chert		Quartzite		Total No.
Block	No.	%	No.	%	No.	%	No.	%	No.	%
1	381	97.9	2	0.5		0.0	6	1.5		0.0	389
2	369	72.9	71	14.0	11	2.2	50	9.9	5	1.0	506
3	149	39.3	45	11.9	26	6.9	139	36.7	20	5.3	379
4	190	85.2	9	4.0	6	2.7	18	8.1		0.0	223
5	235	75.6	13	4.2	15	4.8	36	11.6	12	3.9	311
6	22	39.3	6	10.7	6	10.7	19	33.9	3	5.4	56.
Total	1346	72.2	146	7.8	64	3.4	268	14.4	40	2.1	1864

Table 6-5. Counts of artifactual lithics material types throughout the study region.

The availability of raw material throughout the region is inferred from the variability by material type from artifact collections throughout the study region. In the volcanic region of Block 1 and Block 4, obsidian is the principal locally available material. Block 4 extends into lower elevations regions on the east and west, and chert may be available in streambeds in those regions. Block 2 has larger quantities of fine-grained volcanic stone, mostly andesites, and these were heavily used in Late Archaic. Chert is relatively abundant in Block 2 as well, which suggests that there is a chert source not far from that block. Blocks 3 and 6 show the abundance of chert and chalcedony available in the upper Colca valley area, and multicolor chert cobbles were observed in several stream beds in Block 3. Quartzite is also used in Block 3, a material that was observed eroding out of the ridgetops. Block 5, as with Block 4, appears to contain a variety of raw material types within its boundaries.

Nodules of obsidian had their geological origin entirely in the Maymeja area of Block 1 and the adjacent areas of Block 4. The only other exposure of Chivay type obsidian encountered in the course of this research was the Pulpera / Condorquiña flow - a small exposure of Ob2 obsidian nodules, all measuring less than 5cm, at the toe of a long Barroso lava flow near Pulpera in Block 5. The Condorquiña flow probably saw very little use in prehistory due to small size and poor obsidian quality.

Based on the assumption that the survey was comprehensive and that there are no other exposures of Chivay type obsidian in the Ancachita - Hornillo area, it is proposed that culturally-modified of obsidian was transported radially from the source area in Block 1 and Block 4.

Why quarry for obsidian when it can be found on the surface?

A quarry pit was located on the south side of Block 1 in the Maymeja area of the Chivay source. This quarry will be described below, as well as in Chapter 7 where the quarry pit is examined and the results are presented from a test unit placed in the debris pile associated with the pit.

Further analysis of Ob1 and Ob2 obsidian types at the Chivay source

(a) (b)

Figure 6-4. Photographic comparison of the homogeneous Ob1 obsidian and the Ob2 obsidian with heterogeneities.

Building on the overview of the use of Ob1 and Ob2 obsidian in the preceding section (see Figure 6-3 and Table 6-8 above), further exploration of Ob1 and Ob2 distributions follows. The results show that Ob1 and Ob2 obsidian artifacts in the area of the Chivay source assume patterned distributions over space, and these distributions are probably linked to the use of obsidian for export and for bifacial tool production.

Discussion

The larger patterns revealed by these surface collections can be summarized as follows. First, the Ob1 material appears to have been available in the largest sizes in the Maymeja area of the Chivay source. Second, knappers in the Block 2 area appear to have made greater use of the Ob1 material, but for some reason they had smaller starting nodules as is evident from the smaller artifacts with ? 30% cortex. The fact that cortical materials in the immediate consumption area have much smaller sizes than those being derived from the Chivay source suggests that the largest Ob1 nodules were notbeing consumed in Blocks 2 and 3, and one possible explanation is that they were being exported to the larger consumption region.

We can gain further insights into the differential use of obsidian through the patterns associated with Ob1 and Ob2 material. These data relate to question of the importance of transparent, homogeneous obsidian. In Chapter 3 the relative use of Chivay and Aconcagua obsidian at Asana was discussed because it was inferred that later pastoralists may have been satisfied with Aconcagua material because they were less concerned with the aesthetic qualities of obsidian and more focused on its utility for shearing and butchering. The assumption being that visual quality was less significant for utilitarian applications. For projectile point manufacture, however, Ob1 obsidian appears to have been much preferred by pastoralists. In the pre-pastoral Archaic the use of homogeneous, Ob1 obsidian for projectile point manufacture was less prevalent.

The use of Ob2 can be considered in terms issues of access, aesthetics, and economy.

(1) Access:The obsidian with heterogeneities was found scattered across a larger region on the east and south-east flanks of Hornillo, as well as intermittently on surface elsewhere in the Blocks 1, 4, and 5 survey areas. In contrast, the Block 1 Maymeja area was the only zone with large nodules of Ob1 obsidian available, and under modern conditions the majority of these are beneath a layer of ash. These data suggest that obsidian procurement during the Middle and Late Archaic may have involved more frequent exploitation of surface materials simply because these groups did not have knowledge of, or need to, excavate to obtain Ob1 obsidian. Alternately, during the Terminal Archaic and onwards, quarrying for clear obsidian in the Maymeja zone was developed and greater quantities of clear obsidian were circulating.

(2) Aesthetics:The Ob1 obsidian appears, to modern eyes, that it would have had more value in cultural and prestige related functions. From a biological adaptationist perspective, Ob1 obsidian has higher costly-signaling value (Craig and Aldenderfer In Press), and one would expect both hunter-gatherers and pastoralists to emphasize obsidian free of heterogeneities for its signaling value. As was discussed previously, exchange of objects between individuals or groups as symbolic tokens is documented among hunter-gatherers as well as pastoralists. However, during the pastoralist period social hierarchy increases rapidly and under these circumstances it is possible that the social importance of Ob1 obsidian to a competitive leader during a period of dynamic transegalitarian is considerable.

(3) Economics:Pastoralist producers of projectile points could afford to be selective in the material that they used because projectile points were not necessarily "consumed" by subsistence hunting. If obsidian points are to be used as the principal means meat procurement, points will be broken and lost during hunting forays and there is therefore a need for less costly and easily replaced projectile points. During the pastoralist period, however, hunting for meat is supplementary to the meat available from the herd. Therefore, projectile point use becomes more discretionary because points are used for activities such as non-essential hunting, warfare, or symbolic exchange.

An additional economic component to the use of clear obsidian concerns the economy of projectile point production. Evidently obsidian was predominantly used for Series 5 (concave base, triangular) projectile points styles, and Series 5 points are much smaller on average than other projectile points. When only unbroken projectile points are considered, the mean weight of series 1 through 4 projectile point types is 5.69 g, sd = 4.92, while for Series 5 points the mean weight is 2.08 g, sd = 1.80. On average, Series 5 points are 2.89 times smaller than the other point styles, and therefore one could produce many more projectile points from a single nodule of clear, Ob1 obsidian if one were to make Series 5 points as opposed to an older, larger type of projectile point.

Breakage of projectile points

Obsidian has low compressive strength and the behavior of the material is often characterized as "brittle" (Section 4.4), resulting in a high incidence of breaks in tools that inform as the use of the tool. Latitudinally snapped tips and midsections are typically associated with breakage in use, perhaps upon impact as a projectile. In contrast, broken haft elements, shoulders, and longitudinal breaks are commonly associated with breakage during manufacture or retooling. The evidence for breakage during manufacture is reinforced when the point has incomplete scar coverage because it suggests that the point was discarded during production.

Projectile Points and Material Types through Time

Temporally diagnostic projectile points can be used to look at changes in material type through time in the vicinity of the Chivay source using a time sensitive typology such as the projectile point typology developed by Klink and Aldenderfer (2005). Andeanists have observed that obsidian projectile points were more widely used with the advent of the small, triangular style recognized as belonging to later time periods (Burger, et al. 2000: 294). This style is referred to as the Series 5 point type and it is associated with the Terminal Archaic and onward. It is likely that the frequent use of obsidian for production of the smallest point styles, type 5D, was due to a change in technology, such as the adoption of bow and arrow technology (Klink 2005: 52). This interpretation is supported by the predictable knapping quality of obsidian and the ease with which pressure flaking can be used to produce small points that do not unbalance the arrow in flight, and because the precise pressure flaking also allows resharpening of arrow points with a minimum of loss of material.

Evaluating the entire Upper Colca Survey area, the diagnostic projectile points were found in the following material types. As described in chapter 5, the Series 5 points have not yet been analyzed as closely as the Series 1-4 points because the Series 5 points are not temporally sensitive to the same degree. Therefore Series 5 points are excluded from some tables below where these data are not yet available (as shown in Table 5-9).

Period	Point Type	Obsidian	Volcanics	Chalcedony	Chert	Quartzite
Archaic (General)	3d	489	449, 1019		394
Early Archaic	1a	355, 379, 380, 941, 1015, 1037, 1044.2		384
	1b	398, 411, 493, 522, 945, 956.3, 1026.3	444, 514, 956.2		469, 1034
E. - M. Archaic	2a	512, 949
Middle Archaic	2c	517, 873, 953, 1016, 1023, 1035, 1036, 1044, 1062	822, 1021	790.3
	3b	386, 820, 944	958, 1051		509, 942	951
	3e		519		395
Latter part of M. Archaic		1002
Late Archaic	3f	490	390, 480
	4d		960, 963, 964, 1017, 1018, 1024.4, 1024.5, 1048, 1067	94	38
Late2 - T. Archaic	4f	407, 450, 463, 818, 943, 954, 1031	364, 1057	486
Middle Horizon (?)	Poss. 4e	472

Table 6-13. ArchID numbers for diagnostic projectile points, Series 1-4 only.

Type	Obsidian			Volcanics			Chalcedony			Chert			Total
	No.	mWt	s Wt	No.	mWt	s Wt	No.	mWt	s Wt	No.	mWt	s Wt
1							1	6.00	-				1
1a	7	3.14	1.68										7
1b	7	3.14	1.57	3	4.00	0.00				2	4.50	0.71	12
2a	3	2.67	1.53							1	8.00	-	4
2b	1	1.00	-										1
2c	12	4.55	1.92	3	4.00	1.73	1	3.00	-				16
3a	1	6.00	-	1	4.00	-							2
3b	9	7.78	6.08	2	7.00	0.00				5	4.40	0.89	16
3d	6	3.17	2.04	4	7.50	4.12	1	9.00	-	5	18.00	13.40	16
3e				2	5.50	2.12				1	4.00	-	3
3f	5	3.00	2.00	2	8.00	1.41							7
4d	3	11.00	8.54	11	6.27	3.35	1	4.00	-	1	9.00	-	16
4d / 3b or 3d	9	2.89	1.83	1	3.00	-	1	5.00	-	3	7.00	4.58	14
4e?	1	1.00	-										1
4f	11	3.45	2.54	2	2.50	0.71	1	2.00	-				14
Series 1-4 Totals	75	4.20	3.66	31	5.68	2.91	6	4.83	2.48	18	9.06	8.92	130
5	87	1.63	0.95	2	5.00	2.83	2	6.00	2.83	1	4.00	-	92
5a	6	2.17	1.60							1	7.00	-	7
5b	16	1.63	0.89	1	5.00	-							17
5c	2	2.50	0.71										2
5d	78	1.40	0.54							1	1.00	-	79
Other	35	3.06	2.31	7	4.29	1.98	4	7.25	2.36	5	9.20	7.40	52
Grand Total	299	2.39	2.38	41	5.39	2.73	12	5.83	2.52	26	8.50	8.14	379

Table 6-14. Projectile point mean weights by point type and material type for Upper Colca project. Two quartzite points were excluded.

Table 6-14 shows the weights of all projectile points found on the surface of the project area. It shows that obsidian points were much smaller and much more common in the Series 5 types, and that the few type 5 that are not made from obsidian are relatively heavy. Two quartzite points that were excluded from Table 6-14 include a quartzite point of type 3b that weighted 7g, the other was a probable type 4g (Cipolla 2005) with an excurvate haft and a convex base, and it weighed 14g.

The changing use of a particular material type in association with projectile points is informative in any study region with diagnostic point styles. In the vicinity of an obsidian source these data also have the potential to inform about whether obsidian was used for projectile points out of preference or out of need.

For example, there is a type 3B projectile point, diagnostic to the Middle Archaic, made out of quartzite found in Block 2. This point was found only 16.3 km from the obsidian source, and it was found in a zone rich in obsidian, andesite, and chert, yet the coarsest material in the region was used for producing this point. Heavy materials such as fine-grained volcanics and quartzites known to have been used for projectile when mass rather than sharpness and penetrating power was being prioritized (Ellis 1997), and this is perhaps the explains the use of quartzite in this instance.

	Series 1-4		Series 5		Total
Block	No.	% Column	No.	% Column
1	8	10.67%	11	5.21%	19
2	44	58.67%	141	66.82%	185
3	8	10.67%	19	9.00%	27
4	8	10.67%	30	14.22%	38
5	4	5.33%	9	4.27%	13
6	3	4.0%	1	0.5%	4
Total	75		211		286

Table 6-15. All obsidian projectile points by survey block.

The evidence from all obsidian projectile points (survey and excavation) from the project area shows the prevalence of obsidian projectile points in Block 2. Proportionally, there are relatively few obsidian projectile points in Block 1 which suggests that obsidian was not undergoing advanced reduction in the quarry area. This pattern becomes even stronger in the Series 5 projectile points. While neither Series 1-4 nor Series 5 appear to be involved in advanced reduction in the proximity of the obsidian source, these data suggest that by Terminal Archaic, when Series 5 points were first produced, obsidian acquisition was perhaps more of a special purpose provisioning activity rather than an embedded activity.

Archaic site classifications

Sites occupied by foragers in the Upper Colca survey area were approached using site type classifications based on those used by Aldenderfer (1998: 52-75) in the Osmore drainage. These types of sites include residential bases, logistical camps, hunting blinds, and procurement locations. With very high rates of site reoccupation, discerning the Archaic occupation of any particular site was challenging, and the reoccupation and formation processes of sites strongly impact the older, Archaic component of multicomponent sites. Evaluations were based on the preliminary surface investigations in the course of a larger survey, and therefore the temporal component affiliations and site type assignments presented here should be treated as provisional. Furthermore, many of the sites identified in the Upper Colca were light surface scatters or deflated sites and therefore it is unlikely that archaeological knowledge will become significantly better concerning these sites.

Type	Description	Expectations
Residential Base	- Long term occupation or regular reoccupation by entire families. - Apparently formalized use of space with artifact distributions. - Typically associated with shelter and reliable water source.	Diversity in raw material types and in stages of manufacture in high density reduction loci. Multiple low density lithic scatters apparent throughout site. Artifacts associated with domestic and food preparation activities.
Logistical Camp	Short term but regular reoccupation by special task groups. Spatial location puts a priority on tasks.	Medium-Low assemblage diversity in material types but cores and range of reduction stages evident. Projectile point production failure may appear as tips and midsections in mid to late stage manufacture, and latitudinal snaps. Bases of projectile points from retooling.
Hunting Blind	Small, infrequently used or single-use area.	Low material type diversity. Resharpening and minor retooling.
Procurement and initial production	Associated with raw material source. Frequently exposed or otherwise non-optimal camp location.	High incidence of initial production stages with cortical material. Abandoned cores and decortication flakes.

Table 6-17. Classifications for Archaic components of sites.

These site type classifications were not assigned in the course of field work. Rather, the classifications are a combination of subjective observations during fieldwork by experienced team members and quantitative results from field mapping and from collections and subsequent analysis. Final classifications into site types as portrayed in Table 6-17 occurred in the course of the later analysis of survey data. The environmental characteristics of these site type classifications have been evaluated using GIS data and it is possible to generalize about these site types in comparison with "average" values for each survey block.