Variable Perspectives on “Standardization in the Stone Age”

Overview

Lithic standardization has many definitions, but here we define it as the production of stone tools with limited variation. Yet, many uncertainties, ambiguities, and contradictions remain within such a broad definition. For example, what does it “mean” for toolmakers to standardize their implements? How much variation do archaeologists allow for when considering standardized lithics? Are we talking about the standardization of the final lithic forms as static objects that form the archaeological record, or in using the term standardization, are we talking about inferred processes? What are the implications of lithic standardization regarding human cognition, social behavior, and technological adaptations?

Our aim with this volume was to highlight the importance of lithic standardization as a key, widely used but variably defined and operationalized theme in Paleolithic research. Our goal was to create a space to share various perspectives on lithic standardization from different regional, temporal, and methodological perspectives. Our aim was not to compile a set of papers to probe essential questions about lithic standardization and to use the assembled case studies to consider lithic standardization’s implications for human evolution.

Archaeologists have long discussed the notion of lithic standardization. Mellars (1989), for instance, argued that Eurasian Upper Paleolithic lithic assemblages demonstrated much higher levels of standardization than earlier Middle Paleolithic ones. Yet, others have argued that the African Middle Stone Age shows high levels of standardization (e.g. McBrearty & Brooks, 2000, p. 492), arguing that this level of standardization reflects a critical turning point in human behavioral evolution. Indeed, going even further back in time, others have argued that “there was a general trend of increasing complexity and standardization of stone tools during the Lower Palaeolithic” (Hovers, 2009, p. 9). Indeed, while some archaeologists see tools such as handaxes as representing a kind of conceptual standardization in broad terms, increasing research demonstrates spatially and temporally distinct (standardized?) technologies going back hundreds of thousands of years (e.g. Shipton & White, 2020).

While standardization has outgrown its utility in the minds of many archaeologists, it has also diversified in its interpretive framework. Several frameworks emphasize standardized typologies of retouched tool forms (e.g. Bordes, 1961). These theoretical frameworks led to the over-reliance on formal tool categories, a problem that has continued to evolve over the decades (e.g. McBrearty & Brooks, 2000). In recent decades the increasing emphasis on the dynamism of the reduction process has led to significant conceptual changes in lithic analysis. In approaches such as the chaîne opératoire, the typology of individual forms has, arguably, been replaced by a typology of technologies. Whether the analyst is grouping types of sidescrapers or distinguishing assemblages dominated by the Quina method from those by the recurrent unidirectional convergent method, both reflect concepts of toolmakers standardizing implements enough to fit within one particular category and not another. Yet this is often based on subjective opinions and expert knowledge rather than quantifiable and replicable criteria.

Ideas around standardization are often close to the surface of many prominent debates on lithic technology and human prehistory. For instance, lithic standardization’s implications are inherent in the issues around form and function in the famous “Bordes-Binford” debate. Dibble’s (e.g. 1984, 1995) criticism of desired end products reflects yet another state of the “dynamic” perspectives, which became increasingly prominent in lithic analysis from the late twentieth century. In this view, different “scraper” types were not fundamentally different categories, reflecting distinct mental templates but rather reflecting various stages of shaping and rejuvenation. While conceptually valid, the extent to which researchers have shown such perspectives to be consistent with the archaeological record is debatable. Regardless, retouched tools often comprise only a minor proportion of lithic assemblages. Retouched tools are also conceptually ambiguous, reflecting the trade-off between making a tool a specific shape and resharpening its edges.

Likewise, we could look at standardization from the perspective of cores, but cores are the residual waste product giving information primarily on the final flaking phases. The point here is that all approaches can give partial insights. Still, particular perspectives have weaknesses and strengths, and researchers have struggled to provide broader and deeper-reaching perspectives on standardization and its implications for human evolution. Lithic analysts should celebrate the recent growth of three-dimensional modeling on cores, but the risk of returning to static perspectives, albeit at much higher resolution, should be avoided. We are, or should be, interested in both the final shape of objects and the production processes, but we need help to do both robustly.

While we repeat that our aim was not to offer any final answers on standardization, we suggest some valuable points moving forward. Firstly, archaeologists can and should see standardization from various perspectives (see Table 1). Therefore, researchers need clarity on which perspective standardization is being suggested, be it the core reduction method’s character or the character and location of retouch on flakes. Is a whole artifact being discussed, or rather a particular part? Secondly, expectations should be made clear in advance so that researchers test evaluations of standardization against some benchmark. Finally, while we encourage methodological plurality, basic quantitative reporting, such as using the Coefficient of Variation or other variance tracking measures, is central. The reporting of such measures should be standard in the reporting of lithic assemblages, and this would allow rapid assessments and comparisons of standardization metrics. Researchers have already profitably addressed the notion of standardization as a reduction in variability (e.g. Marks et al., 2001) from the perspective of the Coefficient of Variation (e.g. Eerkens & Bettinger, 2001). Studying lithic standardization, as with many other aspects of lithic analysis, will benefit enormously from a more widespread basic quantification, more sophisticated methods, and the usual call for more samples and better dating.

Table 1. Outline the special edition papers and their definitions and measures of lithic standardization.

Author	Definition of standardization	Measures of standardization
Blessing et al.	Adherence to specific social norms or mental templates	Coefficient of variation
Buchannan et al.	Uniformity in point outline shape	Semi-landmarks and geometric morphometrics
Caruana et al.	Uniformity in handaxe shape and production errors	Average production error indicators, PCA on handaxe measures
Groucutt and Rose	Uniformity in flaking strategies and core shapes	Attribute comparisons and Coefficient of variation comparisons of shape variables
Key and Gowlett	Uniformity in handaxe shape	Coefficient of variation
Khaksar et al.	Uniformity in flake edge morphology	Metric comparisons of stone edge curvature
Li and Lotter	Uniformity in handaxe shape	Metric comparisons of technological and morphological variables
Lombard	Uniformity in measurement systems	Tip Cross Sectional Analysis
Muller and Clarkson	Uniformity in flake morphology	Coefficient of variation
Paige and Perreault	How much certainty or uncertainty does an information system contain	Information Theory
Schoville et al.	Uniformity in stone point morphology	Geometric morphometrics on 2D point images
Shipton et al.	Uniformity in handaxe shape	Geometric morphometrics on 3D scans
Eren et al.	Uniformity in flake edge curvature	Statistical comparisons of means and standard deviations
Shea	The patterned imposition of non-intrinsic shape	Theoretical discussion
Taipale and Rots	The regularization of artifact size and form	Comparisons of tool metrics, raw materials, and use-trace data
Groucutt and Rose	Uniformity in lithic production processes and forms	Coefficient of variation
Wilson et al.	Reduced variability in how toolmakers make stone tools	Statistical comparisons of means and standard deviations

The Papers

The papers in this volume cover a variety of periods, regions, and methodologies (Table 1). Here we briefly introduce the papers, outlining them broadly chronologically, followed by more methodological and experimental approaches.

Wilson and colleagues (2023) explored lithic standardization through the lens of core uniformity in Oldowan flaking systems by conducting knapping experiments, and contrasting different instructional conditions (observation-only and direct active instruction). They found that those knappers that had received direct active instruction flaked more efficiently and with fewer knapping errors leading to differences in core uniformity. Comparing their experimental results to Oldowan cores from Gona, Ethiopia, they found patterns consistent with their experimental results, suggesting an essential role for teaching in the Oldowan. This research combining experimental and archaeological data, highlights the profound relationship between standardization and skill. Wilson and colleagues’ results verify earlier studies by Toth and colleagues (2006), showing that Oldowan toolmakers possessed relatively sophisticated abilities for core reduction and the efficient production of flakes.

Key and Gowlett (2022) also compared experimental and archaeological samples in their study of a replica set of modern handaxes and an assemblage of handaxes from Kenya dating to around one million years ago. They aimed to explore differences and variations in handaxe measurements and shapes between the two assemblages. Interestingly, they found that the replica and archaeological handaxes were similar in size and shape. Still, some differences exist, such as the Kilombe assemblage having allometry not found in the replica set. Key made the study’s replica set as variable as possible within the confines of handaxe technology. The fact that, in most regards, it is as standardized as the study’s various archaeological assemblages raises interesting questions about intentionality and standardization in prehistoric handaxe assemblages. The authors highlight that the study justifies the use of modern replicas in studies of prehistoric lithics.

Li and Lotter (2022) look at the “tongue-like” biface tips from the Baise Basin in southern China. They argue that this handaxe shaping idiosyncrasy represents a deliberate and standardized shaping technique to produce thin, rounded tips. They use experimental knapping studies to suggest that the “tongue-like” morphology was a deliberately imposed strategy to make distal ends that were thin, round, and wide. Highlighting the various ways in which it is possible to address the issue of standardization, with many studies looking at things like assemblage wide variation in shape measures, this study highlights the utility of specific characteristic features of handaxe assemblages. The study also highlights the fact that tools can be standardized in some ways but not in others, such as tip versus overall morphology in this case. These distinctive features form an important part of the record, along with general aspects of shape and technology.

Highlighting the multiple perspectives on handaxe technology, Caruana et al. (2022) explore handaxe shaping and knapping errors from the Late Acheulean site of Amanzi Springs in South Africa. By comparing different phases of the site, they show changing frequencies of aberrant terminations (hinge and step) and cross-sectional shape management (thinning) through time. They interpret these patterns as showing toolmakers’ gradual adaptation to the site’s locally available raw material (quartzite). While researchers often explore standardization by looking at variability in various measures of size and shape, Caruana and colleagues’ study offers insights into the production process from the perspective of knapping errors and how toolmakers dealt with them. Their study shows it is not just what was made that tells us about standardization, but how toolmakers made these implements and what errors they incurred that are key components of artifact standardization.

In the final paper on the Acheulean, Shipton, and colleagues (2023) explore variability in Late Acheulean handaxes from Southwest Asia. Specifically, they compared handaxes from the classic Levantine woodland site of Tabun Cave with recently discovered open-air Acheulean sites from the Nefud Desert of northern Arabia (Khall Amayshan-4 and Khabb Musayyib-1). These assemblages show various technological features in handaxe shape, from highly pointed to highly round. By comparing the three-dimensional morphology of these handaxe assemblages, Shipton and colleagues show that the northern Arabian sites have a low diversity of shapes, in contrast to the highly diverse assemblages from Tabun. They explore possible drivers of handaxe shape and argue that, as well as potential functional differences, the variation may be driven by social factors. The study highlights the utility of three-dimensional approaches to studies of lithic standardization. This paper suggests that the influence of raw material variability could outweigh other factors influencing lithic standardization (with the most variable assemblages made on a single raw material).

Turning to the Middle Stone Age (MSA), Blessing and colleagues (2022) look at the issue of standardization with a case study from Sibhudu Cave, South Africa. The detailed diachronic sequence from this site allows for an understanding of changes in microlithic and blade/let technology. By comparing various aspects of raw material, size, shape, and technological features, they highlight the various forms of standardization, and its absence, in the assemblages compared. They offer an important discussion on the application and utility of the Coefficient of Variation (CV) to study standardization, a measure used by several other studies in this volume. They highlight that the CV values in the 20–50% range fall between those of random production and of precise craft specialists. What, then, are we to make of the meaning of assemblages being relatively, but not highly, standardized and of being standardized in some ways but not others? Whatever the answer to this question, Blessing and colleagues (2022) convincingly argue that researchers should detach the study of lithic standardization from studies of behavioral complexity and symbolism. For instance, they say that in the specific examples they looked at, raw material form played a role in influencing the extent of standardization.

Schoville and colleagues (2023) explore the issue of MSA points, which researchers often discuss as precocious examples of regionally standardized technologies (e.g. McBrearty & Brooks, 2000). They highlight the different perspectives of the hafted part of points and the more visible distal part (the tip) and test the relative standardization of the proximal and distal ends to explore this. They found that, in contrast to their expectations, the proximal ends show greater shape variation than the distal ends. This paper highlights the need to look at standardization in terms of different lithic artifact portions, not just the overall morphology, particularly in the case of hafted technologies. They suggest the results are not consistent with the idea of MSA points signaling group membership and instead highlight the importance of functional aspects (i.e. hafting adaptations).

Groucutt and Rose (2023) explore the standardization of “Nubian” Levallois technology at a site (TH-69) in Dhofar, southern Oman. This site has among the highest representation of Nubian Levallois technology of any site. The site, therefore, forms a key reference point for debates on the meaning and significance of Nubian Levallois technology. While the authors have different perspectives on the wider significance of Nubian Levallois technology (Groucutt, 2020; Rose et al., 2011), their study focuses on the assemblage scale rather than claims for “technocomplex” level discussions. The authors show how in many regards, the assemblage is standardized in terms of reduction methods and artifact shapes, but there is flexibility in how toolmakers achieve this standardization.

Khaksar and colleagues’ (2023) paper uses experimental archaeology to examine the role of flake gross-edge curvature throughout multiple Preferential Levallois Core reduction stages. They test the hypothesis that as gross-edge curvature increases, flake cutting efficiency decreases. Through this project, they tackle the theoretical issue of standardization in Levallois technology and call upon researchers to use Levallois as a theoretical model rather than as a strict, hyper-standardized reality to be recognized. Their results show that Levallois debitage plan- and profile-view gross-edge curvature are largely standardized across reduction stages. They argue that the regularity of Levallois gross-edge curvature may explain a mechanism for Levallois’ widespread occurrences across space and time.

Buchanan and colleagues (2023) look at Middle Paleolithic Levallois points from the East Mediterranean Levant. In this paper, Buchanan and colleagues use geometric morphometric analyses to explore Levallois point shape variation. They then compared Levallois point shape to Folsom points from North America, a terminal Pleistocene point form that researchers regard as highly standardized. They found that Levallois points are less standardized than Folsom points while exploring possibilities to explain this, including how toolmakers made and used them. It is also worth noting that their detailed analysis of shape teases out the variation between Levallois points from different Levantine sites while offering detailed insights into the region’s more traditional typological and chaîne opératoire approaches.

Lombard (2022) addresses the origins of lightweight javelin hunting using a standardized measurement system called the tip cross-sectional area (TCSA) approach. She explores the TCSA approach, which researchers use to hypothesize about distinctions between different stone points and their inferred functions (such as thrusting spears versus projectiles). In this view, different lithic weapon tips show standardized TCSA values according to their implied role (s). Lombard uses this approach to argue for the appearance of lightweight javelin hunting in southern Africa in MIS 6 and perhaps simultaneously in the Levant.

Moving to the more cross-temporal and thematic papers, Muller and Clarkson (2022) offer an important paper that we imagine will be a genre classic. A common theme to emerge in the study of lithic standardization is that researchers need domain-specific benchmarks for what counts as standardization. Muller and Clarkson (2022) establish those benchmarks for lithic technologies. As they discuss, toolmakers can standardize lithic technology in various ways, such as flake production and retouch. They focus here on the basic form, flake production. They explore basic flake morphological standardization in each method by comparing 728 flakes produced using seven lithic reduction strategies (such as discoidal, Levallois, prismatic blade, etc.). As well as providing key information that researchers can compare to archaeological datasets, they highlight a series of three “jumps” in standardization – such as between percussive blade and pressure blade technologies. They argue that toolmakers achieved standardization by increasing skill and investment in core preparation.

While most papers in this volume focus on lithic technology, Taipale and Rots (2023) bring a different perspective with their paper on functional perspectives on lithic standardization. They use use-wear results to explore the influence of tool use mechanics, working edge maintenance, and hafting on lithic standardization. For instance, particular hafting arrangements impose morphological restraints on stone tools. In their paper, Taipale and Rots (2023) compare frequently used standardization measures (such as CV values) in lithic technology with their functional data. They make an important call to integrate functional perspectives into studies of lithic standardization. They also critique the vague ways researchers have used the concept of standardization in their work, which accords with the comments of many of the other papers in the volume.

Paige and Perreault’s (2022) paper provides a primer on information theory using Shea’s (2013) A-I modes framework for lithic classification. The paper asks a question you have probably not thought of but should have, “How surprising are the contents of lithic assemblages?” The authors examine how surprising archaeological finds are given prior knowledge about the archaeological record, how much more we can gain from excavating new assemblages, and how information content can help archaeologists understand taphonomic and cultural biases in lithic assemblages. They point out that lithic datasets’ information content (entropic and syntactic properties) is as important for understanding lithic variability as questions about artifact meaning. They then move to a practical application from these conceptual roots by looking at a database of Late Pleistocene lithic assemblages spanning Africa, Eurasia, and Southeast Asia coded using Shea’s (2013) modes. Their results provide unique insight into the role of information entropy in structuring lithic assemblage standardization.

Finally, Shea (2022) offers a very Shea-like paper. He begins by asking why archeologists expect to find standardization among prehistoric lithics, arguing that it is because most of us grew up in industrialized societies and that, furthermore, various “mirages” create the illusion of standardization. As discussed in other contributions to the volume, these include things like the requirements of hafting and other such pragmatic aspects. For instance, Shea suggests that the apparent relative standardization of large cutting tools may say more about what hominins can realistically carry than about a deep desire to produce standardized shapes. Recognizing lithic “mirages” is crucial in disentangling the different drivers of standardization and may allow the fundamental aspects, such as cognition which drove many early standardization studies, to reappear at a higher level. Shea sums up the widespread concern about a lack of clarity in what we mean by standardization, how we detect standardization in lithic assemblages, and what it “means” if we do find it. Thinking openly about where we are now is, of course, always fundamental in helping guide future directions.

Acknowledgments

We thank Ceri Shipton for his discussions with us and Grant McCall for his role as Editor in Chief of Lithic Technology. We dedicate this special issue to the memory of Donald Henry, a southern Levantine lithic technology specialist, and Bob Brain, a specialist in cave taphonomy, who passed away in 2023.