Developing a stable point: Evaluating the temporal and geographic consistency of Late Prehistoric unnotched triangular point functional design in Midwestern North America

https://doi.org/10.1016/j.jaa.2017.04.001Get rights and content

Highlights

  • The occurrence of unnotched triangular points in the eastern US.

  • Functional–design characteristics of unnotched triangular points from Ohio.

  • Our results show that triangular points may have been designed with hunting in mind.

  • The results also suggest that warfare may have been guiding point design.

Abstract

The occurrence of unnotched triangular points is exceptional in the North American archaeological record. The study of these items can shed light on selective forces that influence the evolution of prehistoric weaponry, especially that which involves small stone tipped projectiles, which is itself a global phenomenon during the late Pleistocene and continuing throughout the Holocene. Following Engelbrecht’s (2015) analysis of points from the Eaton site (1550 CE) in western New York, we assessed potential functional–design characteristics of a large sample of complete and broken unnotched-triangular points from Blain Village, Ohio, an archaeological context several centuries older than, and approximately 550 km southwest of, the Eaton site. We compared our archaeological results to patterns of both experimental point breakage and practical morphometric thresholds for successfully hunting deer. Our results showed that the complete Blain points possessed on average significantly lower length-to-width ratios and significantly higher thickness-to-length ratios relative to the broken ones. Moreover, all points – complete and broken – possessed basal widths under 5 cm and maximum thicknesses under 1.1 cm. These results are consistent with the hypothesis that Late Prehistoric points were designed to effectively penetrate the thoracic cavity of a deer, as well as potentially cause greater trauma during warfare through breakage. Moreover, given that these results are fully consistent with Engelbrecht’s (2015) analysis from the Eaton site, our results indicate that there was a broad, strong selective pressure (i.e., functionally related biases) driving the design stability of unnotched triangular points across the Midwest for an extended period. The principles and methods we use are broadly applicable, especially in terms of using quantitative information derived from experiments to guide our understanding of factors influencing morphometric variation in archaeological points.

Introduction

Stone-tipped projectile weaponry has a long evolutionary history in the archaeological record (Iovita and Sano, 2016, Knetch, 1997, Lombard and Phillipson, 2010, Sahle et al., 2013, Shea, 2006). This extends from its origins in the Pleistocene, through the Holocene, and up to the modern ethnographic record (e.g., Ellis, 1997, Oswalt, 1976). The study of stone projectile points in particular has yielded important insights into human hunting strategies (Frison, 1989, Miller, 2016, Sellet, 2004); production technology and functional/stylistic variability and use (Bergman and Newcomer, 1983, Lipo et al., 2012, Patten, 2005, Pettigrew et al., 2015, Smallwood, 2012, White, 2013); overall composite-weaponry design (Christenson, 1986, Hunzicker, 2008, Hutchings, 2015, Lombard and Pargeter, 2008, Waguespack et al., 2009); mobility and dispersal routes (Bamforth, 2009, Anderson and Gillam, 2000, White, 2014); social and economic organization (Andrefsky, 2006, Buchanan, 2006, Buchanan et al., 2015, Shott and Ballenger, 2007, Shott et al., 2007, Smith and DeWitt, 2016, Whittaker, 1987); and social learning and cultural evolution (e.g., Buchanan and Hamilton, 2009, Buchanan et al., 2014, Eren et al., 2015, Hamilton and Buchanan, 2009, O’Brien et al., 2014, Sholts et al., 2012, Smith et al., 2015).

Understanding sources of variability and stability in projectile point form is clearly an issue within the context of several, if not all, of these discussions. Here, we examine North American triangular, unnotched stone arrowpoints as a case study for understanding more broadly the global phenomenon of small stone-tipped projectile weaponry that occurs during the late Pleistocene and throughout the Holocene. Specifically, we integrate experimental and contextual data with point morphometrics in an attempt to better understand the functional attributes and likely performance characteristics of these triangular, unnotched points. Once these attributes and thresholds are understood within their archaeological context, it becomes possible to assess the selective pressures these items were under over time and across space, as well as potential sources of variation or stability present in the artifactual record.

In recent years, the study of triangular, unnotched stone arrowpoints – a widespread and frequent projectile-tip style in the Late Prehistoric (1000–400 B.P.) period of the North American Great Lakes and Midwest regions (e.g., see Engelbrecht, 2014: Table 1; see also Brush, 2006, Christenson, 1997, Church and Nass, 2002, Fitting and Zurel, 1976, Geier, 1983, Jeske, 1992, Justice, 1987, Kehoe, 1966, Knight, 2005, Kuhn, 1996, Lepper, 2005, McKenzie, 1967, Means, 2002, Oplinger and Prufer, 2006, Peacock, 1986, Prufer, 1967, Prufer, 2006, Prufer and Andors, 1967, Prufer and Pedde, 2006, Railey, 1992, Ritchie, 1971, Schock, 1976, Shane, 1967, Shane and Murphy, 1967, Smith and Crawford, 2002, Stothers, 1976, Stothers and Abel, 2002) – has experienced somewhat of a resurgence (e.g., Bradbury and Richmond, 2004, Bradbury et al., 2011, Bradbury et al., 2012, Carmean, 2009, Cook and Comstock, 2014, Engelbrecht, 2014, Engelbrecht, 2015, Fox, 2015, Pollack et al., 2012). Traditionally, these artifacts, referred to as “Levanna points” or “Madison points” (Perino, 1968, Justice, 1987), have featured prominently in debates on the adoption of the bow and arrow across the continent (e.g., Blitz, 1988, Blitz and Porth, 2013, Shott, 1993). More recently, however, lithic analysts have striven to better understand the extent to which unnotched triangular point technology and morphology varies (Lenadri and Merwin, 2010, Schulenberg, 2010), varies through time (e.g., Church, 2012, Fox, 2015), and whether that variation itself can be used as an independent marker of time (Bradbury et al., 2011, Bradbury et al., 2012, Pollack et al., 2012).

Less attention has been given to the functional implications of unnotched triangular point morphological variation. One exception to this situation is the work of Engelbrecht (2015), who analyzed 516 unbroken points, 103 refitted points, and 808 point bases from the Eaton site in western New York, a multicomponent site with an Iroquoian component comprised of a palisaded village dating to around 1550 CE (400 B.P.) (Fig. 1). With respect to comparisons between the unbroken and refitted points, Engelbrecht (2015) found that the former possessed on average significantly lower length-to-width ratios and significantly higher thickness-to-length ratios relative to the latter. These archaeological breakage results could be logically explained by results stemming from experimental examination of projectile technology. Odell and Cowan (1986:206,208), who fired experimental projectiles at animal targets, found that higher point length:width ratios related to increased penetration and thus greater probability of hitting bone and causing breakage. Alternatively, points with low length:width ratios had a good chance of being deflected. Likewise, Cheshier and Kelly (2006), who shot deer carcasses with obsidian points, found that points with thickness:length ratios greater than 0.121 were significantly more durable than points with ratios lower than this value.

Engelbrecht’s (2015) results have important consequences for our understanding of Late Prehistoric unnotched triangular point functional design. In particular, they are consistent with the hypothesis that unnotched triangular point forms were knapped with both hunting and warfare in mind. With respect to hunting, Engelbrecht (2015:763) notes that the most effective (fastest) means of dispatching deer is by puncturing the lungs, which causes internal bleeding and collapse of the animal. He further posits that if the goal of Late Prehistoric hunters was, therefore, to penetrate the thoracic cavity of a deer, triangular points should possess a basal width of no more than 2.5–5 cm, which corresponds to the distance between the animal’s ribs. Furthermore, extrapolating from the experimental results of Guthrie, 1983, Engelbrecht, 2015 suggested points should be under 1.1 cm in maximum thickness since these have a demonstrably higher success rate in passing between the animal’s ribs compared with points thicker than this value. The vast majority of Eaton site points examined by Engelbrecht (2015) fell within these limits. With respect to warfare, Engelbrecht (2015:763–765) notes that arrows used in battle are not necessarily going to be retrieved. Given that broken projectile points may potentially cause more damage to an enemy than points that remained intact (Ellis, 1997, Frison, 2004), the substantive number of broken points at Eaton with less durable forms raises the intriguing possibility that these points were designed to break upon impact (although more controlled experimental evidence would further support this interpretation). The necessary performance characteristics of both hunting and warfare together drove the evolution of unnotched triangular stone point morphology.

To better understand the temporal and geographic extent of Engelbrecht’s (2015) morphometric results, we examine unnotched triangular points from the Blain Village site, an archaeological context several centuries older than, and approximately 550 km southwest of, the Eaton site (Fig. 1). Such an investigation will reveal whether the functional considerations of point design may have been under strong selective pressure given their expansive breadth in both temporal and regional distribution during the Midwestern Late Prehistoric period.

Following from the discussion above, we were interested in understanding two questions. First, are the length-to-thickness ratios found on both the complete and broken triangular points from Blain Village similar to those found on triangular points from the Eaton site? Second, do the Blain Village points suggest a functional selective pressure related to hunting in that their maximum basal widths are less than 5 cm and their maximum thicknesses are less that 1.1 cm? Given the lack of available lithic artifact refits at Blain, we could not exactly replicate Engelbrecht’s (2015) methods by comparing whole points to refit broken ones. Instead, we compared unbroken and broken points by systematically reconstructing lengths of the proximal point bases by digitally “extending” distally converging lateral edges to the location of intersection, i.e. the “tip”. As described below in Section 2.2.2, we assessed the validity of this procedure via a blind test and adjusted our broken point length estimations accordingly.

Section snippets

Blain Village and the associated unnotched triangular points

Blain Village is located on the west bank of the Scioto River, Ross County, Ohio (Fig. 1). Originally excavated in the late 1960s (Prufer and Shane, 1970), the site represents a single, Late Prehistoric occupation (980–725 B.P.). A prominent feature of the site is “Blain Mound” (Lovejoy and Heiple, 1970), a 46 cm high earthen structure that may have been 20–25 cm taller prior to modern plowing. Excavations at Blain Village uncovered forty-seven features (Prufer and Shane, 1970:15) – namely midden

Unbroken versus broken point morphometrics

The complete points possessed mean length:width ratios of 3.04 and a median of 3.00 (Fig. 8). The broken points meanwhile had mean and median length:width ratios of 4.41 and 4.21 respectively. Moreover, these length:width ratios were significantly different for complete versus broken points (Uneq. Var. t = −8.44, p < 0.0001; Mann W. U = 1135, Monte Carlo p < 0.0001), suggesting the latter may have had greater penetrability thus having greater probability of hitting bone, as well as being weaker in

Discussion

The occurrence of unnotched triangular points is exceptional in the North American archaeological record. While certain point types (e.g., Clovis) or point features (e.g., side notches) may have had as large, or larger, geographic coverage as did unnotched triangular points, the sheer frequency and density of the latter artifact type is striking. Hence, understanding how, and for what potential purposes, this set of material culture evolved has implications for better understanding a major

Acknowledgements

M.R.B. is financially supported by the Kent State University Biomedical Sciences (Biological Anthropology) Program. M.I.E. is supported by the College of Arts and Sciences, Kent State University. S.J.L. is supported by is supported by the Research Foundation for the State University of New York.

References (97)

  • S. Sholts et al.

    Flake scar patterns of Clovis points analyzed with a new digital morphometrics approach: evidence for direct transmission of technological knowledge across early North America

    J. Archaeol. Sci.

    (2012)
  • D. Anderson et al.

    Paleoindian colonization of the Americas: implications from an examination of physiography, demography and artifact distribution

    Am. Antiq.

    (2000)
  • W. Andrefsky

    Experimental and archaeological verification of an index of retouch for hafted bifaces

    Am. Antiq.

    (2006)
  • C. Bergman et al.

    Flint arrowhead breakage: examples from Ksar Akil, Lebanon

    J. Field Archaeol.

    (1983)
  • J. Blitz

    Adoption of the bow in prehistoric North America

    North Am. Archaeol.

    (1988)
  • J. Blitz et al.

    Social complexity and the bow in the Eastern Woodlands

    Evol. Anthropol.

    (2013)
  • A. Bradbury et al.

    A preliminary examination of quantitative methods for classifying small triangular points from Late Prehistoric sites: a case study from the Ohio River Valley

    Midcontinental J. Archaeol.

    (2004)
  • A. Bradbury et al.

    Kentucky’s small triangular subtypes: old theories and new data

    J. Kentucky Archaeol.

    (2011)
  • A. Bradbury et al.

    Points on points: a reply to Pollack

    J. Kentucky Archaeol.

    (2012)
  • N. Brush

    Rockshelters in the Lower Killbuck Valley, Holmes and Coshocton Counties, Ohio: 1982–88

  • B. Buchanan et al.

    A formal test of the origin of variation in North American early Paleoindian projectile points

    Am. Antiq.

    (2009)
  • B. Buchanan et al.

    Continent-wide or region-specific? A geometric morphometrics-based assessment of variation in Clovis point shape

    Archaeol. Anthropol. Sci.

    (2014)
  • K. Carmean

    Points in time: assessing a Fort Ancient triangular projectile point typology

    J. Anthropol. Archaeol.

    (2009)
  • J. Cheshier et al.

    Projectile point shape and durability: the effect of thickness: length

    Am. Antiq.

    (2006)
  • Christenson, A., 1981. The Evolution of Subsistence in the Prehistoric Midwestern United States. Unpublished Ph.D....
  • A. Christenson

    Projectile point size and projectile aerodynamics: an exploratory study

    Plains Anthropol.

    (1986)
  • A. Christenson

    Side-notched and unnotched arrowpoints: assessing functional differences

  • Church, W., 2012. Temporal shifts in lithic technology at the Clark Site (33Wa124): A Late Woodland site in Southwest...
  • Church, F., Nass, J., Jr., 2002. Central Ohio Valley during the Late Prehistoric period: subsistence-settlement...
  • R. Cook et al.

    Toward more continuous and practical artifact analyses: defining and learning from key dimensions of Fort Ancient triangular projectile points in the Miami Valleys

    Midcontinental J. Archaeol.

    (2014)
  • C. Ellis

    Factors influencing the use of stone projectile tips: an ethnographic perspective

  • W. Engelbrecht

    Unnotched triangular points on village sites

    Am. Antiq.

    (2014)
  • W. Engelbrecht

    Interpreting broken arrow points

    Am. Antiq.

    (2015)
  • M.I. Eren et al.

    Test, Model, and method validation: the role of experimental stone artifact replication in hypothesis-driven archaeology

    Ethnoarchaeology

    (2016)
  • Fitting, J., Zurel, R., 1976. The Detroit and St. Clair River Area. In: Brose, D. (Ed.), The Prehistory of the Lake...
  • Fox, A., 2015. A study of Late Woodland projectile point typology in New York using elliptical Fourier outline...
  • G. Frison

    Experimental use of Clovis weaponry and tools on African elephants

    Am. Antiq.

    (1989)
  • G. Frison

    Survival by Hunting: Prehistoric Human Predators and Animal Prey

    (2004)
  • C. Geier

    Some thoughts on the taxonomic and temporal implications of Late Woodland triangular projectile points from sites in the Ridge and Valley Province of Virginia

    Archaeol. Soc. Virgina Quart. Bull.

    (1983)
  • Guthrie, R., 1983. Osseous projectile points: biological considerations affecting raw material selection and design...
  • Ø. Hammer et al.

    PAST-palaeontological statistics, ver. 1.89

    Palaeontol. Electron.

    (2001)
  • J. Hart

    Monongahela subsistence-settlement change: the late prehistoric period in the lower Upper Ohio River Valley

    J. World Prehistory

    (1993)
  • D. Hunzicker

    Folsom projectile technology: an experiment in design, effectiveness, and efficiency

    Plains Anthropol.

    (2008)
  • R. Iovita et al.

    Multidisciplinary Approaches to the Study of Stone Age Weaponry

    (2016)
  • R. Jeske

    Energetic efficiency and lithic technology: an Upper Mississippian example

    Am. Antiq.

    (1992)
  • N. Justice

    Stone Age Spear and Arrow Points of the Midcontinental and Eastern United States

    (1987)
  • T. Kehoe

    The small side-notched point system of the Northern Plains

    Am. Antiq.

    (1966)
  • A. Key et al.

    Edge angle as a variably influential factor in flake cutting efficiency: an experimental investigation of its relationship with tool size and loading

    Archaeometry

    (2015)
  • Cited by (23)

    • Stone tool backing and adhesion in hunting weaponry: First results of an experimental program

      2022, Journal of Archaeological Science: Reports
      Citation Excerpt :

      While certainly possible, such a conclusion in turn presumes that increased adhesion and decreased shaft splitting would have been desired traits. If, however, past peoples designed their projectile weaponry to “fail” for the purposes of causing more harm to prey such that dislodged backed segments slice and damage tissues (e.g., Bebber et al., 2017; Ellis, 1997; Engelbrecht, 2015), then the cultural evolutionary implication of our results changes. Indeed, this latter possibility is consistent with the notion that backed tools were the equivalent of “disposable razors” (Ambrose, 2010, p. 141) and potentially explains why Lombard (2007) found quartz segments at Umhlatuzana hafted with a brittle single component adhesive recipe designed to shatter on impact.

    • Experimental assessment of lanceolate projectile point and haft robustness

      2022, Journal of Archaeological Science: Reports
      Citation Excerpt :

      Like other performance characteristics–such as penetration, aerodynamics, sharpness, and multifunctionality–point robustness (the ability to withstand breakage) was an attribute either consciously or unconsciously manipulated by prehistoric peoples in specific settings (Buchanan and Hamilton, 2020; Eren et al., 2021; Iovita et al., 2014; Pargeter 2007; Wilkins et al., 2012). In some contexts, stone weapon tips may have been designed to resist breakage (Thomas et al., 2017), in other contexts foragers may have created points to increase the chances of fracturing and therefore causing more damage (Bebber et al., 2017; Engelbrecht, 2015; Mika et al., 2020). Here, we focus on the former, situations where the goal was to produce robust points capable of withstanding breakage.

    • Modern thermoplastic (hot glue) versus organic-based adhesives and haft bond failure rate in experimental prehistoric ballistics

      2021, International Journal of Adhesion and Adhesives
      Citation Excerpt :

      Likewise, poor hafting bond failure rates should also not automatically be viewed as a “negative”. In the same way a broken stone projectile tip may increase damage to prey or an enemy [5,6,61], a stone tip that becomes detached from its shaft may produce additional internal damage to prey or an enemy, while allowing the shaft to be retrieved. One must also acknowledge, however, that not everyone in the past had ready access to the same resources.

    • Controlled ballistics tests of ground, percussion-flaked, and pressure-flaked projectile point impact durability: Implications for archaeological method and theory

      2019, Journal of Archaeological Science: Reports
      Citation Excerpt :

      This latter result should be explored further in future tests, but we suggest that this result has less to do with production technique differences (i.e. percussion vs. pressure flaking), and more to do with the morphometric differences between the samples. Point morphometrics has been demonstrated numerous times archaeologically and experimentally to strongly influence point breakage (e.g. Bebber et al., 2017; Cheshier and Kelly, 2006; Engelbrecht, 2015; Odell and Cowan, 1986). Along these lines, Table 1 shows that the percussion flaked and pressure flaked sample comparisons exhibited the most number of significant mean differences in terms of form, and that these mean differences possessed moderate to large size effects.

    • Wound ballistics: The prey specific implications of penetrating trauma injuries from osseous, flaked stone, and composite inset microblade projectiles during the Pleistocene/Holocene transition, Alaska U.S.A.

      2018, Journal of Archaeological Science
      Citation Excerpt :

      Flaked stone points had the highest rebound rates both because of failure to penetrate the hide (9 rebounds into soft tissues) and minimal penetration when then hitting bone (7 rebounds into hard tissues). These rebound rates are characteristic and in-line with other experiments on points with low length to width ratios (Bebber et al., 2017; Odell and Cowan, 1986), and likely resulted from increased bone interference - preventing clean passes between the ribs (Pettigrew et al., 2015; Sisk and Shea, 2009). Since the rib spacing of large ungulates can vary dramatically, point width is an important technological metric to consider if seeking to increase the odds of a lethal penetration (Friis-hansen, 1990).

    View all citing articles on Scopus
    View full text