Sex, symmetry and silliness in the bifacial world.

Hayden, Brian ; Villeneuve, Suzanne

After 10 years of pursuing sexy handaxes it is probably time to put these coquettish creatures to bed. Readers wishing to continue the debate are courteously directed to our Project Gallery

As lithic technologists, we were at first amused by Kohn and Mithen's (1999) arguments to the effect that handaxes developed in order to enhance sexual selection among early hominins. However, contributors to Antiquity continue to take their suggestions seriously and even to extend them to new heights (Machin 2008; Mithen 2008; Hodgson 2009). We would like to bring the issues back down to the ground, and discuss the practical and technological aspects of biface manufacture which these authors do not seem to have appreciated. This is perhaps not surprising since none of the authors appear to have much experience in manufacturing stone tools. We do agree with them that a socially-oriented analysis of stone tools is desirable, but this must be founded on a solid understanding of lithic technology.

Kohn and Mithen's (1999: 520) original argument, as with subsequent versions published in Antiquity, is based on the fundamental assertion that 'in the majority of artefacts a specific symmetrical form was imposed ... an imposed symmetry beyond functional requirements.' This assertion was supported by a study of the effectiveness of symmetrical versus asymmetrical handaxe performance in butchering (Machin et al. 2006). Mithen (2008: 766-9) continues to claim that his sexual selection explanation is the only theory that can account for the features typical of handaxes. One hesitates to accept these claims at face value when also confronted with highly questionable assertions to the effect that bones were used as billets and handaxes were used for chopping up vegetables. The same fundamental premise has been uncritically adopted by subsequent commentators who propose variously that consistency of form in handaxes was the product of natural selection (for unstated reasons--Machin 2008: 763); or that symmetry evolved among animals in general as a means of identifying things in the world and that handaxe symmetry evolved to create reassurance that an unspecified critical aspect of the human perceptual system was working appropriately (Hodgson 2009). Machin subsequently presented cogent arguments from a Darwinian viewpoint as to why handaxes would make poor candidates for use in sexual selection.

What Kohn and Mithen fail to acknowledge is that there is a practical technological logic behind the symmetry and other characteristics of bifaces. This does not appear to be very evident from cognitive evolutionary perspectives. What we suggest is that analysing handaxe symmetry from a design theory viewpoint can be much more insightful. Design theory involves: 1) the identification of a specific problem to be solved; 2) the identification of relevant constraints (typically factors such as material costs and availability, performance characteristics, required skill, processing volumes, time availability, existing technology, mobility and transport limitations, and available labour); and 3) identifying the trade-offs in advantages and disadvantages of certain constraints (such as material procurement costs) against the advantages and disadvantages in other constraints (such as effectiveness). These factors are then used to develop a suite of design solutions (Pye 1964, 1968; Horsfall 1987; Hayden 1998; Hayden et al. 2000).

[FIGURE 1 OMITTED]

In order to demonstrate the potential utility of design theory, the issue of handaxe symmetry can be considered in the broader context of general trends throughout the Stone Age toward greater economy in the use of lithic material employed in high-consumption activities. Greater material economy is manifested by changes in core reduction techniques and resharpening techniques; and these, in turn, arguably reflect changes in major design constraints. Oldowan choppers/cores, for example, produce a minimal number of flakes or resharpenings per kilogram of tool stone due to the hard hammer techniques employed. The large striking platform remnants that result rapidly reduce the volume of core material and increase edge angles (Figure 1a & b). Handaxes, on the other hand, are made with billets which enable knappers to resharpen tools many more times and to remove many more flakes from a given mass of raw material due to the much smaller striking platform remnants in relation to flake sizes (Figure 1c). Reduction of cores using blade techniques extends the amount of cutting edge per kilogram of material even further (Sheets & Muto 1972) as does the production and use of microliths. Pressure flaking, edge grinding, and the use of resharpenable metals all extend the lives of cutting tools into the realms of curated tools that can last for years or even generations. Each of the technological developments mentioned exhibits symmetry, requires skill, training, and often special raw materials or auxiliary tools that go far beyond the simple cutting requirements of daily routines. Are these developments all to be viewed as sexual or costly signalling rather than basic technological adaptations to underlying material and performance constraints?

[FIGURE 2 OMITTED]

We suggest that the handaxe can be viewed as a specialised tool design ultimately responding to increased processing and time constraints. Torrence (1989) and Bleed (1986) have argued that time constraints (the need to complete certain activities within a limited amount of time) may explain why specialised tools were developed. Under conditions of time stress, such as the need to butcher and dry thousands of salmon during short spawning runs, or the need to butcher and remove meat from large game before packs of predators arrive, one cannot afford to interrupt harvesting or cutting activities because of tool exhaustion and wander off in search of materials (often several kilometres distant) to replace what has been used up. From a design theory perspective, it can also be argued that as processing volumes increase, the development of specialised tools becomes more cost-effective even though they usually involve increased procurement/manufacturing costs, skills, and special manufacturing equipment (Hayden & Gargett 1988; Hayden 1989). This is a relationship that has been well-established by engineers for industrial machinery (Figure 2; see also Bell 1972; Zeyher 1977; Gold 1979) but has also been documented in technologies as simple as fish butchering knives (Figure 3; see also Hayden & Gargett 1988). And it was most likely processing volume constraints that were changing most significantly throughout the Palaeolithic and Neolithic (Hayden 1989). At this basic level, the same fundamental technological relationships may be as applicable to the Palaeolithic as they are to recent centuries.

During the Palaeolithic and Neolithic, the greater the processing volumes the more inefficient it would have been to discontinue a suite of processing activities in order to obtain replacement raw materials for exhausted tools. In addition, like the industrial lathes and fish butchering knives in Figures 2 and 3, the greater the processing volumes were, the more cost effective it would have been to increase the efficiency of the tool by creating a specialised tool that would perform tasks with maximal efficiency and maximal longevity even though it might be more costly to make or maintain. These relationships especially obtain when the transport of lithic material is constrained by human mobility, i.e. the need for individuals to carry all of their tools and replacement materials with themselves during moves.

[FIGURE 3 OMITTED]

To return to the Palaeolithic, the scenario of increasing cutting requirements over time seems especially plausible if we assume that, during Oldowan times, the use of spears may have been relatively infrequent and limited (Hayden 2008). Chopping tools and the flakes derived from them appear to have been entirely adequate to cope with rare butchering or spear making/sharpening events between suitable lithic sources. With the advent of Acheulean technology, however, it is generally thought that hunting became increasingly perfected and that hunted meat played a much larger role in the human diet (Shipman & Walker 1989; Aiello & Wheeler 1995; Bramble & Lieberman 2004). If so, this implies that spears and butchering tools were being made and used far more frequently than in the Oldowan, and involved much greater processing volumes of both wood (for spears) and meat. Given these new processing requirements, together with the constraints of transporting all tools by hand, it would have been important to develop more specialised tools and to make those tools last significantly longer than Oldowan choppers. It is in this context that the development of handaxes makes good technological sense, despite the increased costs of training, making and transporting specialised billets, and procuring better quality raw materials.

As illustrated in Figure 1c, billet-produced bifaces can be resharpened many more times than any core or core tool reduced by hard-hammer techniques. Moreover, the resulting billet flakes typically have very acute edge angles making them ideal for cutting through hides of animals. Such tools dull quickly when cutting through dirty or muddy hair and they may need to be replaced frequently, thereby making production of these flakes in quantity an important consideration in some contexts. In this respect, one of the most important functions of handaxes in some situations may have been as sources of thin billet flakes as well as bifacial butchering tools. It seems evident that handaxes would have made very useful tools for removing skins and for other aspects of butchering, as demonstrated in a number of experiments and through use-wear analyses (Jones 1980; Keeley 1980: 169, 170, 175; Schick & Toth 1993; Machin et al. 2006).

If this much can be established with some certainty, it is critical to realise that the much-touted symmetry of handaxes is also a product of the soft-hammer reduction technique. In effect, it makes no technological sense to use billets to reduce lithic materials unless one is trying to maximise the number of resharpenings or to produce thin billet flakes. In both cases, this can only be systematically and reliably accomplished by creating a symmetrical, lenticular cross-sectioned core tool. In fact, the more lenticular and symmetrical the form, the easier it is to remove typical thin, flat, curved, tear-drop shaped billet flakes that travel beyond the centre of the tool and hence maintain the cross-sectional and edge angle morphology within constant ratios or optimal ranges over the many flake removals typical of soft-hammer reduction (Hayden & Hutchings 1989). As any accomplished flint knapper knows, this is what underlies the symmetry of handaxes whether people were trying to maximise the use-life (number of resharpenings) of handaxes as tools, or trying to maximise the number of sharp flakes removed from a core tool, or both. We argue that the symmetrical and lenticular shape of general handaxes had no more to do with their use in sexual selection than the symmetrical rounded shape of a pencil point over the many resharpenings of its lifetime. As the adage states: form follows function. We suggest that this is also the reason why billet reduction everywhere in the world, whether Europe, Africa, or the Americas, exhibits symmetrical, lenticular bifacial forms--or do Kohn and Mithen and others want to argue that all bifaces everywhere in the world were used as sexual selection or other symbolical strategies?

Far from being the only theory that explains handaxe symmetry or other characteristics, Kohn and Mithen's scenario appears as a weak supposition lacking grounding in technological realities. Handaxes do not even make reliable indicators of reproductive merit since they could be easily loaned out and may have circulated more as communal property than individual property, as is typical of almost all material possessions in simple hunter-gatherer societies. Kohn and Mithen's suggestion strains credulity that handaxes were discarded shortly after being made (sometimes in abundance) in order to show off skills by flint knappers who were being watched by potential mates (Kohn & Mithen 1999: 522). This idea is not much better than the informal suggestion that some archaeologists have expressed that handaxe makers lacked memory to the extent that when they rested, they put down their bifaces and then forgot that they had already made a handaxe and so started over, and over, to make yet another.

It is far more plausible to suggest either that some localities were being used as depots to stockpile tools like handaxes and raw materials for easy access during transits that did not intersect good quarry locations (much as suggested by Potts (1988: 305) for Oldowan lithic accumulations), or that the main objective of handaxe production at some sites was not the handaxes themselves but the production of quantities of billet flakes for skinning or making bags or cloaks of skin. This could account for the relatively visually 'pristine' appearance of many handaxes at some well-documented butchering sites like Boxgrove, and the large numbers of plain, unretouched flakes at some sites (the thin edges of most billet flakes are not very suitable for direct percussion retouching, and even when resharpened will never have as sharp an edge as an the unretouched version).

Nor can much credence be given to the claim that 'Changes in sexual selection criteria ... caused the Acheulean ta break dawn' (Kohn & Mithen 1999: 523). It is far more likely that changes in hunting, woodworking, hafting, or other economic constraints, were the underlying factors leading to the abandonment of handaxe technology. Curated, hafted Levallois flakes, in particular, may have replaced expedient billet flakes as the skinning tool of choice in many areas.

All this is not to deny that objects requiring some skill to make could have sometimes been used to impress others, especially when well made--a point also made by Machin (2008: 765-6). But this is true of virtually all crafts, including some lithic types. It seems almost certain that same bifaces in later periods were used as prestige items. Good examples include the very rare and exceptionally large and thin, non-functional Solutrean laurel leafs, and the ceremonial bifaces (up to 750mm long) used by California Indians in rituals (Kroeber 1905: 690, 1953: 26-7; Rust 1905). However, in these cases, the prestige or ritual examples were all exceptional variants of basic utilitarian bifaces. While there are good ethnographic accounts of elaborate bifaces being used to impress ritual specialists or high ranking individuals, there are no ethnographic indications that bifaces were ever used as a significant strategy to attract mates, or for that matter to project any 'important perceptual imperative' related to symmetry or other abstractions or to trigger any sense of reassurance that the perceptual system was working appropriately (Hodgson 2009: 196-7).

That a few handaxes, especially the very large examples, may have been used to impress others, or even used as non-functional prestige objects, is a possibility worth exploring and might provide insights into Acheulean social dynamics. However, deriving the existence of entire technological classes of objects on the basis of the extreme characteristics of a few rare and exceptional cases is not sound methodology. It is no more appropriate to derive the reason for handaxe development from a few large examples than it is to try to argue that automobiles in general developed primarily as a sexual selection or prestige display tactic on the basis of the elongated and lavish attributes of stretch limousines.

The basic symmetrical biface form makes far more sense as firmly rooted in the practicalities of lithic inventorying, mobility, processing volumes, and other fundamental design constraints. In order to explain the appearance and demise of specific archaeological aspects, it may be trendy to attempt to clothe old artefacts in the mantles of new theories like sexual selection, signalling theory, or evolutionary psychology of symmetry. However, without a firmly grounded understanding of the basic technology involved and its constituent constraints, it soon becomes apparent that the new clothes of such theoretical emperors are embarrassingly lacking in substance.

Postscript

This comment complements the recent article by Nowell and Chang (2009).

Acknowledgements

We would like to express our deep gratitude for the lithic insights and flint knapping training that Francois Bordes provided, the suggestions from Maxine Kleindienst to pursue design theory applications in lithic studies, and the cogent comments provided by Rick Schulting on an earlier draft.

References

AIELLO, L. & P. WHEELER. 1995. The expensive-tissue hypothesis. Current Anthropology 36: 199-221.

BELL, R. 1972. Changing technology and manpower requirements in the engineering industry (Science Policy Research Unit, Research Report 3). Brighton: Sussex University Press.

BLEED, P. 1986. The optimal design of hunting weapons: maintainability or reliability. American Antiquity 51: 737-47.

BRAMBLE, D. & D. LIEBERMAN. 2004. Endurance running and the evolution of Homo. Nature 432: 345-52.

GOLD, B. 1979. Productivity, technology and capital Lexington (MA): Heath.

HAYDEN, B. 1989. From chopper to Celt: the evolution of resharpening techniques, in R. Torrence (ed.) Time, energy and stone tools: 7-16. Cambridge: Cambridge University Press.

--1998. Practical and prestige technologies. Journal of Archaeological Method and Theory 5: 1-55.

--2008. What were they doing in the Oldowan? Lithic Technology 33: 105-39.

HAYDEN, B. & R. GARGETT. 1988. Specialization in the Paleolithic. Lithic Technology 17(1): 12-18.

HAYDEN, B. & K. HUTCHINGS. 1989. Whither the billet flake?, in D. Amick & R. Mauldin (ed.) Experiments in lithic technology (British Archaeological Reports International Series 528): 235-57. Oxford: British Archaeological Reports.

HAYDEN, B., N. FRANCO & J. SPAFFORD. 2000. Keatley Creek lithic strategies and design, in B. Hayden (ed.) The ancient past of Keatley Creek (Volume 1): 185-212. Burnaby (B.C.): Archaeology Press.

HODGSON, D. 2009. Symmetry and humans: reply to Mithen's 'sexy handaxe theory.' Antiquity 83: 195-8.

HORSFALL, G. 1987. Design theory and grinding stones, in B. Hayden (ed.) Lithic studies among the contemporary highland Maya: 332-77. Tucson (AZ): University of Arizona Press.

JONES, P. 1980. Experimental butchery with modern stone tools and its relevance for Palaeolithic archaeology. World Archaeology 12: 153-65.

KEELEY, L. 1980. Experimental determination of stone tool uses: a microwear analysis. Chicago (IL): University of Chicago Press.

KOHN, M. & S. MITHEN. 1999. Handaxes: products of sexual selection? Antiquity 73:518-26.

KROEBER, A. 1905. Notes by A.L. Kroeber. American Anthropologist 7: 690-95.

--1953. Handbook of the Indians of California. Berkeley (CA): California Book Co.

MACHIN, A. 2008. Why handaxes just aren't that sexy: a response to Kohn & Mithen. Antiquity 82: 761-6.

MACHIN, A., R. HOSHELD & S. MITHEN. 2006. Why are some handaxes symmetrical? Testing the influence of handaxe morphology on butchery effectiveness. Journal of Archaeological Science 34: 883-93.

MITHEN, S. 2008. 'Whatever turns you on': a response to Anna Machin, 'Why handaxes just aren't that sexy'. Antiquity 82: 766-9.

NOWELL, A. & M.L. CHANG. 2009. The case against sexual selection as an explanation of handaxe morphology. PaleoAnthropology 2009: 77-88.

POTTS, R. 1988. Early hominid activities at Olduvai. New York: Aldine de Gruyter.

PYE, D. 1964. The nature of design. London: Studio Vista.

--1968. The nature of art of workmanship. Cambridge: Cambridge University Press.

RUST, H. 1905. The obsidian blades of California. American Anthropologist 7: 688-9.

SCHICK, K. & N. TOTH. 1993. Making silent stones speak: human evolution and the dawn of technology. New York: Simon & Schuster.

SHEETS, P. & G. MUTO. 1972. Pressure blades and total cutting edge: an experiment in lithic technology. Science 175: 632-4.

SHIPMAN, P. & A. WALKER. 1989. The costs of becoming a predator. Journal of Human Evolution 18: 373-92.

TORRENCE, R. 1989. Re-tooling: towards a behavioral theory of stone tools, in R. Torrence (ed.) Time, energy and stone tools: 57-66. Cambridge: Cambridge University Press.

ZEYHER, L. 1977. Zeyher's new guide to cost reduction in plant operations. Englewood Cliffs (NJ): Prentice-Hall.

Brian Hayden & Suzanne Villeneuve *

* Archaeology Department, Simon Fraser University Burnaby, British Columbia, Canada V5A 1S6 (Email: bhayden@sfu.ca)