Research Approach

We hold, as other researchers have (Urry, Reference Urry2007; Greenblatt, Reference Greenblatt and Greenblatt2010; Frankopan, Reference Frankopan2015; Marshall, Reference Marshall2015), that examination of the specific, categorical and quantifiable conditions surrounding human movement can be highly informative. Many useful attempts have been made to describe the social structure of mobile groups along a continuum (e.g. Beardsley et al., Reference Beardsley, Holder, Krieger, Meggers, Rinaldo and Kutsche1955; Murdock, Reference Murdock1967; Binford, Reference Binford1980). Other approaches (Kelly, Reference Kelly1992, Reference Kelly1995) involve the quantification of residential moves per year by average distance covered. However, it is arguable that, ‘the reality of a mobile existence is far more complex than the ordering principles used to describe it’ (Wendrich & Barnard, Reference Wendrich, Barnard, Barnard and Wendrich2008: 15). It seems that the most accurate means of modelling mobility is through a thorough examination of a variety of phenomena in combination with archaeological context. Notable examples of these defining criteria include:

1. 1. Mobility (length of time, season);

2. 2. Number of journeys;

3. 3. Segment of the population which moved (as defined by gender, age, health, occupation, or social position);

4. 4. General socio-political organization;

5. 5. Logistics and available modes of transport.

In an ideal world, these fiveFootnote ¹ categories (see discussion, e.g. Urry, Reference Urry2007; Wendrich & Barnard, Reference Wendrich, Barnard, Barnard and Wendrich2008; Cabana & Clark, Reference Cabana, Clark, Cabana and Clark2011; Adey et al., Reference Adey, Bissell, Hannam, Merriman and Sheller2014) should be investigated via multiple samples from multiple individuals from a site, region, and culture group who represent the full gamut of ages, sexes, and social levels. Unfortunately, the fragmentary nature of the archaeological record rarely includes material suitable for covering all parameters.

While both individual and group mobility data are inherently very different (as are mobility and migration), historically, archaeology has examined them in somewhat similar ways. Archaeologists have used both migration and mobility as catch-all categorizations for the causes of cultural change. The roots of this style of interpretation date back to culture-historical approaches and the drawing of Kulturkreise (‘culture spheres’; Frobenius, Reference Frobenius1898, Reference Frobenius1899) which characterized material culture as a proxy for language, ethnicity, and self-identity. However, the increasingly rigorous typologies and chronologies also made ‘les autres’ (‘the others’; Tordorov, Reference Tordorov1992) more apparent in terms of the style and manner of the items with which they were buried (Jockenhövel, Reference Jockenhövel1980, Reference Jockenhövel1991, Reference Jockenhövel and Jockenhövel1995; Wels-Weyrauch, Reference Wels-Weyrauch1989; Jockenhövel & Kurbach, Reference Jockenhövel and Kurbach1994).

In recent decades, advances in mobility research have created a broader understanding of how mobility—or potential mobility—functions in an increasingly connected world (see discussion of push/pull in Heberle, Reference Heberle1938; Lee, Reference Lee1966; cost/benefit in Sjastaad, Reference Sjastaad1962; and ‘motility’ in Kaufmann et al., Reference Kaufmann, Bergman and Joye2004). New methods in the field of isotopic tracing, such as the analysis of hair and fingernails, micro-sampling of multiple skeletal elements as well as the consecutive analysis of molars, has enabled scholars to create high-resolution timelines for individual human movements (Font et al., Reference Font, van der Peijl, van Wetten, Vroon, van der Wagt and Davies2012; Tipple et al., Reference Tipple, Chau, Chesson, Diego, Fernandez and Ehleringer2013; Knipper et al., Reference Knipper, Meyer, Jacobi, Roth, Fecher and Stephan2014; Frei et al., Reference Frei, Kristiansen, Allentoft, Wilson and Skals2015a, Reference Frei, Mannering, Douglas Price and Birch Iversen2015b, Reference Frei, Villa, Jørkov, Allentoft, Kaul and Ethelberg2017; Scharlotta, Reference Scharlotta2018). Studies of various cases of single individual mobility timelines can be combined with other group analyses, thereby allowing archaeology to gain a better handle on the complexity of the networks connecting human biology, ethnicity, culture, and heritage (Haak et al., Reference Haak, Brandt, de Jong, Meyer, Ganslmeier and Heyd2008; Knipper et al., Reference Knipper, Mittnik, Massy, Kociumaka, Kucukkalipci and Maus2017). In short, the course of academic dialogue is changing from producing evidence for movement to examining how different human mobilities either resemble or differ from each other across temporal and geographical barriers. Moreover, we are beginning to provide new and unprecedented information that can help us grasp why that movement first came about.

A Mobility Model

Thirty years ago, David Anthony criticized archaeologists for their approach to migration: ‘instead of developing the needed tools, archaeologists have avoided the subject’ (Anthony, Reference Anthony1990: 895). The increased temporal depth and breadth of mobility research coupled with better baseline coverage and new osteological techniques (e.g. Evans et al., Reference Evans, Montgomery, Wildman and Boulton2010; Frei, Reference Frei2013; Eerkens et al., Reference Eerkens, Barford, Jorgenson and Peake2014; Kootker et al., Reference Kootker, van Lanen, Kars and Davies2016; Frei et al., Reference Frei, Villa, Jørkov, Allentoft, Kaul and Ethelberg2017; Stewart et al., Reference Stewart, Gerlach, Gowland, Gron and Montgomery2017; Hoogewerff et al., Reference Hoogewerff, Reimann, Ueckermann, Frei, Frei and van Aswegen2019) have allowed archaeologists to address the five criteria for mobility assessment in an increasingly concrete fashion. Although there are (and always will be) holes in the record, we propose a mobility model composed of four over-arching mobility patterns which we have named as follows: 1) non-migratory; 2) point-to-point migratory; 3) back-and-forth; 4) repeated mobility, subdivided into 4a (cyclical mobility) and 4b (non-cyclical mobility) (see Table 2).

Table 2.

Mobility model: the schema proposes a series of five different patterns of human mobility.

The following consists of a series of short discussions exemplifying these four kinds of mobility patterns from the present (mainly European) data. Given the preliminary nature of this model and the vagaries of interpretation, we have intentionally produced broad categories containing overlapping elements. We anticipate that the model will require later re-evaluation as more data appear.

Non-mobility: The Case of Singen (Germany)

Within this type of mobility pattern, individuals are stationary in terms of migration and mobility. They tend to be born, live, and die within the same area or region. We propose that a stationary farmer would exemplify this pattern. Within archaeological contexts, this type of pattern is visible in individuals whose burial context and isotopic data both indicate local provenance without apparent signs of mobility. It also includes individuals whose isotopic values in different human tissues point to a local origin despite the presence of non-local archaeological associated material(s), other atypical styles/decorations, and/or manners of construction within a specific grave.

An example of the latter can be seen near Lake Constance at the cemetery of Singen (Hohentwiel). Although the site contains elements of Corded Ware, Bell Beaker, Urnfield, and La Tène cultures (Kimmig, Reference Kimmig and Krämer1958), Singen is best known for its Early Bronze Age finds (Krause, Reference Krause1988). While typological parallels for some of the objects (e.g. the daggers) have been found as far away as southwest Britain and in Brittany, Singen is particularly well known for the kind of metal used in the production of Early Bronze Age artefacts. The characteristic copper alloys used in what has become known as ‘Singen metal’ have been found along the Baltic and the Carpathian Basin (Krause, Reference Krause1988; Oelze et al., Reference Oelze, Nehlich and Richards2011). While the evidence for such far-reaching connections might lead one to expect a high proportion of non-locals at the site, strontium and oxygen isotope analyses conducted by Oelze et al. (Reference Oelze, Nehlich and Richards2011) suggest that the individuals they examined were overwhelmingly local. Even though the people from Singen clearly had contact with the outside world, in most cases, the mobility pattern exhibited by the inhabitants was non-migratory. Hence, this serves as an example of cultural exchange through means other than migration, where individuals seem to be local despite of the ‘non-local’ ancillary contextual evidence at the site. As studies of additional cemeteries within the region and the period as a whole come to fruition, and when additional samples—or perhaps even sequential samples from otherwise apparently ‘non-mobile’ people—become available, scholars may be able to shed more light on whether Singen represents a singular case or whether its non-mobility data forms part of a wider pattern.

Point-to-Point Mobility: The Case of the Skrydstrup Woman (Denmark) and the Capacocha Child Sacrifice in the Andes

A point-to-point mobility pattern can be distinguished from non-mobility through the single, unidirectional movement of the individual(s) concerned. Here we provide examples of individuals who appear to have moved either because of exogamy or pilgrimage/sacrifice. However, many other causes could result in this kind of mobility pattern, e.g. victims of warfare, famine/natural disaster or the need/desire to relocate for professional reasons.

An example of such type of mobility can be seen in the detailed mobility investigations of one of Denmark's well-preserved oak coffin burials: that the Skrydstrup Woman. The Skrydstrup female burial offered a unique opportunity for a detailed mobility study because the human remains consist of skeletal tissues as well as soft tissues like hair (Broholm & Hald, Reference Broholm and Hald1939). The Skrydstrup female was unearthed in 1935 from a burial mound dated to the Early Nordic Bronze Age period III (1300–1100 bc) (Frei et al., Reference Frei, Villa, Jørkov, Allentoft, Kaul and Ethelberg2017). Until recently, scholars assumed that she had local geographical origins because she was buried in the manner of the local elite. Frei and colleagues (Reference Frei, Villa, Jørkov, Allentoft, Kaul and Ethelberg2017) recently analysed dental enamel from her first and third molar as well as sequential hair segments (Skrydstrup Woman's total scalp hair length was over 60 cm) and were able to create a detailed timeline for her movements. All in all, the strontium isotope results indicate that Skrydstrup Woman moved from one geological area (represented by the signature from her first and third molars as well as part of her scalp hair) to another (represented by the remainder of her hair samples and a lower strontium isotopic signature in the third molar compared to the first molar). The detailed baseline study of the area combined with these values suggests that she most probably lived locally at the second location (Skrydstrup) until the end of her days. These results seem to imply that the Skrydstrup female might have been an example of what Bronze Age literature refers to as the ‘fremde Frau’ (‘foreign woman’; Jockenhövel, Reference Jockenhövel1991, Reference Jockenhövel and Jockenhövel1995), i.e. a woman who moved from one area to another as a means of creating and/or validating an economic-political alliance or kinship bonds between different—but related—social groups (Wels-Weyrauch, Reference Wels-Weyrauch1989; Bergerbrant, Reference Bergerbrant2007).

Regardless of the reason(s) why she moved, the results from the new analysis of Skrydstrup Woman suggest a single, unidirectional movement. In other words, a single migration. As this mobility coincided with Skrydstrup's early adolescence, exogamy is a potential cause for her individual mobility pattern. However, alternative scenarios might include unstable political/environmental situations, movement related to kinship networks or perhaps even adaptations to changes in the long-distance networks related to the metal trade (Ling et al., Reference Ling, Hjärthner-Holdar, Grandin, Billström and Persson2013, Reference Ling, Stos-Gale, Grandin, Billström, Hjärthner-Holdar and Persson2014; Melheim et al., Reference Melheim, Grandin, Persson, Billström, Stos-Gale and Ling2018).

Within specific contexts, we can expand our understanding of single point-to-point mobility to include a sacrificial element. If we move our focus from Europe to the Incan Empire, we find another interesting example of unidirectional movement. The capacocha rituals of the Inca involved the sacrifice of children and material goods to the gods as a way of alleviating the results of natural disasters or of honouring key events in the life of the Inca emperor (Reinhard & Ceruti, Reference Reinhard and Ceruti2015). The rituals included the selection of materials and young persons from various regions, their transport to Cuzco, and then further movements to places of particular ritual importance throughout the empire (Ceruti, Reference Ceruti2004). Against a backdrop of written sources that describe the practice of sourcing sacrificial victims for the capacocha ritual from throughout the Inca empire, Andrushko et al. (Reference Andrushko, Buzon, Gibaja, McEwan, Simonetti and Creaser2011) confirmed the presence of non-local child sacrifices by strontium isotope analysis of the children's human remains. The authors concluded that some of the children travelled or were brought to Cuzco Valley with the specific intent of being sacrificed. This conclusion had already been posited for the mummies from the Sara Sara (Peru) and Llullaillaco (Chile/Argentina) volcanoes in Wilson and colleagues’ (Reference Wilson, Taylor, Ceruti, Chavez, Reinhard and Grimes2007) examination of carbon, nitrogen, oxygen, and sulphur isotopes. These studies provide evidence for unidirectional travel ending in the demise of the victims.

Back-and-Forth Mobility: The Case of Haraldskær Woman (Denmark)

Back-and-forth mobility can be distinguished from a point-to-point mobility pattern by the simple fact that, in the former case, the mobile individual returns to their starting point, whereas, in the latter, they do not. A case study of back-and-forth mobility comes from the Danish pre-Roman Iron Age (500–0 bc) bog body of Haraldskær Woman. Recovered by peat cutters working near Vejle in Jutland in 1835 (van der Sanden, Reference van der Sanden1996), the body, which also became known as Queen Gunhild (a mistaken attribution to the Viking Age), is a well-preserved bog body (Glob, Reference Glob1969). Recent detailed strontium isotope analyses of the body's tissues (tooth enamel and scalp hair) and the wool textile recovered with the remains revealed a ‘there-and-back-again’ mobility pattern (Frei et al., Reference Frei, Mannering, Douglas Price and Birch Iversen2015b). While the ⁸⁷Sr/⁸⁶Sr ratios of Haraldskær Woman's tooth enamel suggest that she was likely to have spent her childhood within the area of present-day Denmark, the strontium isotope ratios from her scalp hair suggest that she undertook a long-distance journey shortly prior to her death. The authors suggest that she may have been on a kind of pilgrimage which could have been related to her end as an offering in the bog (Frei et al., Reference Frei, Mannering, Douglas Price and Birch Iversen2015b).

Repeated mobility

We have grouped together different kinds of repeated mobility in a single over-arching group. As a general rule, what we have termed ‘repeated mobility’ involves movements over an individual's lifespan between two (or more) areas. While there are several different forms that this kind of mobility can take, we have sub-divided the category into cyclical and non-cyclical mobilities.