Clovis First / Pre-Clovis Problem
Revisited 2004

Tony Baker
September 15, 2004



Almost seven years have passed since I wrote The Clovis First / Pre-Clovis Problem and Bruce Bradley added his comments. In the intervening years I have studied Clovis material from the Gault Site, the Mockingbird Gap Site, and the Little River Clovis Complex. I have participated in excavations of a Solutrean Site in France and a Magdalenian Site in Spain. Finally, for the last five summers, I have performed survey work for the BLM on the North Slope of Alaska. My knowledge of Clovis and the associated time period has increased significantly, and I now feel the need to write a follow-up. During this same period, the number of Clovis sites and C-14 dates have increased very little. So, the following is based on logic and speculation. I am sure many readers will disagree with my new ideas as they did with the ideas in the 1997 page. Still, the 1997 page was and still is one of my most popular webpages. I hope this one approaches it in interest.

Anagenesis vs. Cladogenesis Evolution
Anagenesis is "directional evolution of a character, combination of characters, or a species within a lineage without diversification, or branching, of the lineage" (O'Brien and Lyman 2000:397).

Cladogenesis is "the branching and diversification of lineages during evolution. It involves the multiplication of taxa within a lineage" (O'Brien and Lyman 2000:398).

Yes, I am aware that anagenesis and cladogenesis are biological concepts and projectiles do not breed. Still, in this webpage I have chosen to apply these concepts to projectile type change through time.

Thin-bodied projectiles are associated with split-stick hafting and are plano-plano in cross-section or have flat faces. (Not to be confused with the Llano/Plano usage.) Flakes scars run beyond the mid-line, if the projectiles are not fluted. Width-to-thickness ratios are greater than 4.0.

Thick-bodied projectiles are associated with socket hafting and are convex-convex or lenticular-to-diamond shape in cross-section. Flake scars stop at the mid-line. W/T ratios are less that 3.0. W/T ratios between 3.0 and 4.0 are in a gray area.

I borrowed these terms from Bob Patten.

When I was a teenager in the 1960's, my father and Wormington's Ancient Man in North America indoctrinated me with the anagenetic model of Paleoindian, projectile type evolution. I do not know if Wormington intended to present an anagenetic model or not, however, she did imply it with the temporal organization of her book. Additionally, her drawings in the back, strongly reinforced this model. In Table 1, I have reproduced her sequence, which was the archaeological understanding at that time.
Table 1
Anagenetic Models
Agate Basin
Agate Basin
Hell Gap

I was still a strong believer in the anagenetic model in 1991. The archaeological understanding had changed a little, but not the model as can be seen in Frison's Prehistoric Hunters of the High Plains. Again, refer to Table 1. In 1999, Dixon began to mix the anagenetic and cladogenetic models in his Bones, Boats, & Bison. He had Clovis disappear as it evolved into Folsom and Goshen. Subsequently, Folsom and Goshen disappear as they evolve into the several types of the Plano or late Paleoindian (1999:152). But, from a different perspective, he held fast to the anagenetic model. Clovis, Folsom and Goshen are thin-bodied projectiles. The projectiles of the Plano are thick-bodied. Dixon anagentically evolved thin-bodied projectiles into thick-bodied projectiles at the beginning of the Plano.

In 1996, I received my first introduction to cladogenetic model in an article by Eighmy and LaBelle concerning pooling and calibrating of radiocarbon dates for various projectile types. Their article demonstrated that the dates for Agate Basin/Hell Gap, Clovis, Cody, Folsom, and Plainview could overlap in time (1996:59). Granted there was just a small portion of overlap of the probability curves of Clovis and Cody, but there was still an overlap. As for Clovis and Folsom, their probability curves had more than 50% overlap. If one accepted the mathematics, then it was obvious that some of these projectile types (traditions) were contemporaneous. Unfortunately, at that time I would not let the facts get in the way of my deeply engrained anagenetic model.

In early July of this year, my anagenetic model fell apart. It didn't fall apart because of any single fact, but from the burden of inconsistencies. One was the re-examination of the Hell Gap excavation by Sellet. In his dissertation work he found that Agate Basin projectiles were stratigraphically above and below Folsom projectiles (1999:118-119). Why did it take me five years to realize the significance? I can only make excuses.

Figure 1
NW Alaska
max. width = 35.7 mm
max.thickness = 12.7 mm
Figure 2
NW Alaska
max. width = 35.3 mm
max.thickness = 12.3 mm
Another inconsistency that troubled me was the C-14 dates associated with thick-bodied, lanceolate projectiles from North and South America. In 1998, the antiquity of a thick-bodied projectile from Alaska was established at the Irwin Sluiceway site (Rasic 2003:24). See Figures 1 and 2. (Click on the images for larger view.) Since that time, a number of additional Sluiceway sites have been identified in Alaska and Sluiceway appears to have a time depth of circa 10,000 to 11,200 rcybp. El Jobo is a thick-bodied lanceolate projectile found in Venezuela and Northern Brazil. At the Taima-Taima site, El Jobo projectiles were found in the body cavity of a mastodon and the C-14 dates from Taima-Taima are older than Clovis (Bryan and Gruhn 1979:53). To put the Sluiceway and El Jobo dates into perspective with Clovis and Agate Basin, another thick-bodied projectile, see Figure 3. Figure 3 depicts all the dates associated with Agate Basin, Clovis, El Jobo and Sluiceway that I can find in the literature. The few not presented are very old, El Jobo dates from Taima-Taima that would compress the vertical scale too much, if they were presented. The vertical bars represent one standard deviation around the mean calendar date. There is a 68% chance that the actual date lies within the range of the bar. Generally, longer bars represent dates acquired before the AMS process was developed. I have converted the C-14 dates to calendar dates with the CALIB REV4.4.2 (Stuiver and Reimer 1993:215-230). A detailed listing of the C-14 dates by sample number are available in the Appendix.

Figure 3

C-14 dating is not the last word in chronology, but it is one of the best. There is also stratigraphy, seriation, and other independent, direct dating methods. My complaint against C-14 dates is the way archaeologists "cherry pick" the data. How many dates have never been published because they did not match the investigator's anagenetic model? How many published dates have been scrutinized and attacked because they were inconsistent with the anagenetic model? Or, how many bad dates have been embraced and never challenged because they were consistent with the model? That said and taking Figure 3 at face value, it suggests a chronological overlap of the Clovis culture by a group of thick-bodied, lanceolate projectiles.

Complex projectiles such as Clovis, which are adopted by a large group of people, are not designed one evening and put into service the next day. There must be a Pre-Clovis projectile that logically evolved into Clovis. Since, the 1960's I have been asking "How does a non-fluted, stone projectile evolve into a fluted, stone projectile and leave no transitional evidence in the archaeological record?" The answer for me was, "that it did leave evidence, but it has not yet be found." In 1997, while waiting for the evidence to be found, I proposed a Solutrean-Clovis connection in my original Clovis First / Pre-Clovis Problem in an attempt to answer the curious similarity of two lithic technologies. The connection was weak and unsatisfying. However, at the time it was the best I could do with the paradigm of the anagenetic model that Pre-Clovis was yet to be discovered.

In July of this year, I finally crossed the Rubicon and exchanged my anagenetic model for a cladogenetic model. Now I can accept the concept that a thick-bodied projectile existed at the same time as Clovis. The two were contemporary on the North American Continent and there is a strong possibility that the thick-bodied projectile was the archetype. Now I can ask, "What is the evidence for a non-fluted, stone projectile evolving into a fluted, stone projectile?" Unlike the question associated with my anagenetic model, which assumed the evidence did not yet exist, this question assumes it does.

The Evidence for a Pre-Clovis/Clovis Transition
Figure 4
Agate Basin or Old Clovis
New Mexico
max. width = 26.2 mm
max.thickness = 8.5 mm
Figure 5
As it turns out, the evidence for a Pre-Clovis/Clovis transition has been in my possession for almost as long as the wrong anagenetic model. Figures 4 depicts a proximal fragment of a projectile, which is one of six, that I have classified as Old Clovis in the past. These six projectiles were the inspiration for my Old Clovis/New Clovis dichotomy that I presented in my The Clovis/Folsom Transition webpage. My father and I found these projectiles, and those that I classify as New Clovis, on a single surface site in the 1960's. This Site, 13JB5, was the only Clovis site cited in Paleoindian Occupation of the Central Rio Grande Valley in New Mexico (Judge 1973:73-74).

Today, I believe the Old Clovis fragment in Figure 4 and the other five are actually refurbished, thick-bodied projectiles. Since Agate Basin is the nearest in time and space to Clovis in the current archaeological understanding, I will call them refurbished Agate Basins. I believe the projectile in Figure 4 began its life as an Agate Basin. It was subsequently broken and became a distal fragment. Next this distal fragment was refurbished by thinning the broken proximal edge, which created a fluted base or at least a fluted base to my former anagenetic eyes. This re-based projectile was again used for hunting and was again broken. At that time it was lost to the archaeological record until we found it in the 1960's.

Figure 5 depicts the same Old Clovis fragment next to an Agate Basin from the Texas Panhandle. I created this image to help me convey my concept of the Agate Basin - Old Clovis evolution. I have positioned the fragment next to the complete projectile where I believe it was located in the original Agate Basin. The Old Clovis has a maximum width and thickness of 26.2 mm and 8.5 mm. The measurements of the Agate Basin 29.7 mm and 8.7 mm.

In summary, I am suggesting that Old Clovis, as defined in The Clovis/Folsom Transition, is actually a refurbished thick-bodied projectile. I will term this thick-bodied projectile as Agate Basin for convenience. From these basally thinned, thick-bodied projectiles, full length fluting (dome and planing) gradually developed and New Clovis evolved. This evolution was cladogenetic because New Clovis split off from Agate Basin while Agate Basin continued to exist during and after Clovis times.

Follow-up and Closing Comments
As alluded to earlier, the anagenetic evolutionary model for Paleoindian has severe seriation problems. The most obvious is, where is the thin-bodied ancestor to the fluted thin-bodied projectile we know as Clovis? The subsequent Clovis/Folsom transition is easier to accept because it is thin-bodied, fluted projectile to thin-bodied, fluted projectile. However, the Folsom to Plano "... transition is more complex than the proceeding period when fluted points were in vogue" (Dixon 1999:228). I will suggest it is much closer to illogical. This would be a case of a thin-bodied, fluted projectile evolving into several thick-body, unfluted projectiles.

The cladogenetic model eliminates many of these seriation problems. There are thick-bodied projectiles, which are associated with microblades, in Upper-Paleolithic of Siberia. Therefore, there is an ancestral source for thick-bodied points in the New World. For example at Dyuktai Cave they are dated between 12,100 +120 and 13,200 +250 rcybp (Mochanov and Fedoseeva 1996:165&169). I propose that in the New World some thick-bodied, unfluted projectiles split off and evolved into thick-bodied, fluted projectiles, which I have termed Old Clovis. Others remained thick-bodied Agate Basin-like. Old Clovis evolved into thin-bodied New Clovis and then into thin-bodied Folsom. (Again, see my Clovis/Folsom Transition webpage.) Finally, Folsom disappeared, but Agate Basin and other thick-bodied projectiles did not. They became the later Paleoindian types.

Big blades are a trait commonly associated with Clovis. They are also a trait associated with lithic rich regions. According to Collins: "Because of limitations in the properties of stone, the major blade-producing technologies of the world are centered on abundant sources of superior chert or obsidian"(1999:12). So, as to the source of Clovis' blade making, they either inherited it from a Pre-Clovis tradition, or they developed it after the Pre-Clovis/Clovis transition. If they inherited it from Pre-Clovis, then the Pre-Clovis/Clovis transition had to occur in a lithic rich region. On the other hand, if the transition occurred in a lithic poor region, then Pre-Clovis would not have been making and using blades and, therefore, Clovis could not have inherited the trait. Using this logic, I find it difficult to accept that Pre-Clovis was necessarily a blade maker.

Bone tools and red ochre are also commonly cited as Clovis traits. They are also common in the Upper Paleolithic throughout the Old World, including Siberia and Alaska, where Clovis did not reside. I suggest that when these materials were available they were used. And all peoples used them. The tales we weave about them are probably more a function of archaeological preservation than cultural traits. Similar to blades, I see no reason to argue for bone tools or ochre as indicators of cultural connectivity.

Consider the thousands of Clovis projectiles that have been found in the 48 states. With a cladogenetic evolution, I would expect a similar number of thick-bodied projectiles to have been found. At least a number within a magnitude of the number of Clovis projectiles. The skeptical reader is probably asking "where are they?" I suggest they are everywhere. How many basally thinned, thick-bodied projectiles have been wrongly identified as Clovis? Or, how many are there that are not branded with a flute and instead are burdened with the anagenetic model, which suggests they are recent? Consider the only "Clovis" projectile found at the Sheaman Site. It has a width-to-thickness ratio of 3.8. Frison writes, it "... demonstrates a slightly atypical Clovis manufacture technology in that the flutes have been obliterated to some extent by flaking subsequent to their removal (1982:152-153). From Frison's images, I can interpret it to be a refurbished Agate Basin. Also, from the same site, Frison's Figure 2.91b is a complete biface, about which Frison writes "... is probably a projectile point preform ..." (1982:147-148). Again, I can interpret it to be an Agate Basin.

Figure 6
The ideas presented here are not new for me. They have been boiling like uncooked potatoes in a stew pot for a number of years. The Mesa projectile of Alaska is a Paleoindian projectile. Based on a vast number of C-14 dates circa 10,000 B.P., "the simplest scenario is that the Mesa complex represents a later, post-Clovis phenomenon, a result of a northward migration following deglaciation" (Bever 2001:101). This simple scenario is the anagenetic model. A more complicated scenario, using the cladogenetic model, is that Mesa has a much deeper time depth. Supporting an older Mesa time frame are a few 11,000+ C-14 dates from the Mesa and the Tuluaq Sites (Bever 2001:101). However, in my own mind, the real support for the deeper time depth is my personal experience on the North Slope. I believe that the Mesa projectile is only a refurbished Sluiceway and Sluiceway does have the time depth. See Figure 3.

"Technologically the Mesa (Sluiceway/Mesa) assemblage appears closely related to the Agate Basin complex of the North American high plains" (Kunz and Reanier 1996:503). I have argued in this paper that Agate Basin is ancestral to Old Clovis. So it follows that Sluiceway/Mesa is ancestral to Clovis. Kunz and Reanier (1996:503) indirectly alluded to this.

A couple of years ago, long before I adopted the cladogenetic model, I noticed a similarity between basally thinned Mesa projectiles and my own Old Clovis projectiles. So, I created a composite of images that I had at the time. (Unfortunately, they have a blue tint to them, but they still show the flake scars.) I emailed the composite to a few Internet groups for discussion. Figure 6 is that composite and the Mesa projectiles are on the right. The third image down on the left is the projectile in Figure 4.

To close, I want to recall a discussion between Al Goodyear and myself a couple of years ago. Al indicated that he also had recognized the Old Clovis/New Clovis dichotomy (thick-bodied/thin-bodied projectiles). However, he disagreed with my chronology. He believed the thick-bodied Clovis projectiles were later in time than the thin-bodied ones. I asked him why. He said in the Southeast, thick-bodied, unfluted projectiles follow Clovis and it made more sense for thick-bodied Clovis to evolve into thick-bodied unfluted projectiles. Ironically, this is the same logic I use for the thin-bodied Clovis being the more recent. In the West, Folsom, a thin-bodied projectile, follows Clovis. To me it made more sense for a thin-bodied projectile to evolve into a thin-bodied Folsom. With my new cladogenetic model, we both can be correct.

Click to load the Appendix -- C14 dates.

Click to load the Postscript of 9/28/04.


Bever, M. R.
1996  Stone Tool Technology and the Mesa Complex: Developing a Framework of Alaskan Paleoindian Prehistory. Arctic Anthropology. 38(2):98-118.
Bryan A. L. and R. Gruhn
1979  The Radiocarbon Dates of Taima-Taima. In Taima-Taima:A Late Pleistocene Paleo-Indian Kill Site in Northermost South America, Final Reports of 1976 Excavations, edited by C. Ochsenius and R Gruhn, pp. 53-58. South American Quaternary Documentation Program, Federal Republic of Germany.
Collins, M. B. 1999
1999  Clovis Blade Technology. The University of Texas Press, Austin.
Dixon, E. J.
1999  Bones, Boats, & Bison. The University of New Mexico Press, Albuquerque.
Eighmy, J. L. and J. M. LaBelle
1996  Radiocarbon Dating of Twenty-Seven Plains Complexes and Phases. Plains Anthropologist 41:53-69.
Frison, G. C.
1982  The Sheaman Site: A Clovis Component. In The Agate Basin Site: A Record of the Paleoindian Occupation of the Northwestern High Plains., edited by G. C. Frison and D. J. Stanford. Academic Press, New York.
1991  Prehistoric Hunters of the High Plains, 2nd edition. Academic Press, New York.
Haynes, G
2002  The Early Settlement of North America. University Press, Cambridge.
Jaimes, A.
1998  El Vano, Venezuela: El Jobo Traditions in a Megathere Kill Site. Current Research in the Pleistocene 15:25-26.
Judge. W. J.
1973  Paleoindian Occupation of the Central Rio Grande Valley in New Mexico. University of New Mexico Press, Albuquerque.
LaBelle, J. M.
2004  Hunter-gatherer foraging variability during the Early Holocene of the Central Plains of North America. Unpublished PhD dissertation, Department of Anthropology, Southern Methodist University, Dallas.
Mochanov, Y. A. and S. A. Fedoseeva
1996  Dyuktai Cave. In American Beginnings: The Prehistory and Palaeoecology of Beringia, edited by F. H. West, pp. 164-174. The University of Chicago Press, Chicago.
O'Brien, M. J. and R. L. Lyman
2000  Applying Evolutionary Archaeology: A Systematic Approach. Kluwer Academic/Plenum Publishers, New York.
Rasic, J. T.
2003  Ancient Hunters of the Western Brooks Range: Integrating Research and Cultural Resource Management. In Alaska Park Science: Connections to Natural and Cultural Resouce Studies in Alaska's National Parks. USDI National Park Service, Anchorage.
Sellet, F.
1999  A Dynamic View of Paleoindian Assemblages at the Hell Gap Site, Wyoming: Reconstructing Lithic Technological Systems. PhD Dissertation, Department of Anthropology, Southern Methodist University, University Microfilms, Ann Arbor.
Stuiver, M., and Reimer, P. J.
1993  Extended 14C Database and Revised CALIB Radiocarbon Calibration Program. Radiocarbon 35:215-230.
Wormington, H. M.
1964  Ancient Man in North America, 5th edition. Denver Museum of Natural History, Denver.

Return to Paleoindian & Other Archaeological Stuff

eXTReMe Tracker