Metonymy and metaphor

Metonymy is normally understood as a linguistic figure of speech, but like metaphor (cf. Tilley Reference Tilley1999) it also operates at a broader, more conceptual level. Simply put, metonymy is at play whenever a sign points to something else based on an existing relationship (cf. Lakoff & Johnson Reference Lakoff and Johnson2003; Radden & Kövecses Reference Radden, Kövecses, Panther and Radden1999). Our everyday speech is full of metonymy, in a way that is so pervasive that it goes unnoticed. Archaeology students go on a dig; some have a good eye for context changes. When we use metonymy it works to draw attention to a particular quality of the thing associated. In these examples we see how archaeology is strongly linked with digging, and excavation technique with visual acuity. In some cases other metonyms could be used—one might have a good feel for different contexts, emphasizing the haptic elements of excavation rather than the visual, but in others, the association is so dominant that another metonym cannot be substituted. In emphasizing some elements (digging) metonyms also obscure or de-emphasize other dimensions (pot-washing and wheel-barrowing). (If you announced you were going on a ‘wheel-barrowing’ you’d be met with blank incomprehension.) Metonymy can therefore not only be used to evoke something complex, but also to restrict how it is understood, by focusing narrowly on one feature. In directing attention towards some things and away from others it can be a powerful tool for shaping how political discourse and identity is understood.

Lakoff and Turner suggest that that metonymic mappings generally remain within the same conceptual domain, in contrast to metaphors, which work to build connections across domains (Lakoff & Turner Reference Lakoff and Turner1989, 103; see also Tilley Reference Tilley1999, 5). Metonymy stays close, while metaphor jumps about. This means that ‘digging’ as a metonym must always stay connected conceptually to the practice of digging itself, and in that way remains attached to the practical work of archaeology. We can contrast this with digging as metaphor, which takes us out of archaeology and into the realm of psychoanalysis, police investigations, or journalism as part of the broader conceptual domain of digging down as a way to imagine the work of enquiry and analysis (Kuspit Reference Kuspit, Gamwell and Wells1989; Thomas Reference Thomas2004). And yet, as Ahmed discusses (Reference Ahmed2004), metonymy can be used to fasten together disparate elements and create firm and reiterated connections that were perhaps latent or not foregrounded previously. Digging as an activity can connect archaeologists with builders, farmers, gardeners and engineers. It might remind us of childhood pleasures and holiday sandcastles, or natural disasters and the rescue efforts they entail. This then can allow metonyms to move more widely than might be at first thought, and also suggests that the conceptualization of domains might imply more coherence than is there in practice. It is helpful here to turn to William Croft, who refined Lakoff and Turner’s proposal by suggesting that metonymy occurs within a single domain matrix which ‘possesses a unity that is created by experience’ (Reference Croft and Peeters2000, 231). This ability to shift through a complex series of connected fields and to bring them into a variegated and extensible domain that is not fully recognized or even graspable by all participants seems to be important in terms of metonymy’s political and affective stickiness.

How a domain is imagined and articulated depends on the experience that is brought to it—in this sense there are no simple boundaries, and domains themselves are attached to other domains through links of association. Metonyms dwell within the connections between things. What interests me here is how metonymy can evoke a feeling, yet at the same time it can also underpin more discursively formulated arguments. When we speak of a dig, we indicate that archaeological research is primarily about excavation, whether or not this is the case. The metonym does not make an explicit claim for truth or falsity, but instead participates in a feeling for what is important and valued in archaeology. This is not a feeling that can be described as an emotion; rather it lies in the realm of affect, as a feeling that is indeterminate, and usually unreflected upon. Its salience emerges from familiarity with the things connected: the connection feels right because it aligns with past experience. If, after Ahmed, we consider the metonymic associations that shape today’s political affects, it is clear that these connections feel true to those who act upon them, yet they do not operate within the realm of fact-checking and evidential proofs. Metonymic signs, the things they point to and the feelings that emerge around these connections are bundled together in a way that feels immediately present and is difficult to disentangle. We should carefully consider how this may also have played out in the past.

Metonymy is an instance of the semeiotic category of indexicality, which Carl Knappett (Reference Knappett2005) has explored in relation to material culture. He shows how artifacts evoke meanings indexically from their associations with other entities, through contiguity, causality and factorality (or part–whole relations). Many archaeologists are probably already familiar with an example of indexical associations through material culture in Madagascar, in relation to the carvings found on Zafimaniry houses in the southern highlands. In a short article that has become well cited and influential in archaeology (Boivin Reference Boivin2009; Garrow Reference Garrow2012; Jones Reference Jones2007; Whittle Reference Whittle1997, to mention a few), Maurice Bloch pushed back against the impulse to reduce everything to representational meaning, noting that the beautiful incised carvings found on traditional houses are ‘pictures of nothing’, carved to make the wood more beautiful and to honour it (Bloch Reference Bloch, Hodder, Shanks and Alexandri1995, 213–14). In another article I have explored this perspective in more detail in relation to the positive affect associated with the carvings and how they index a wider set of changing associations (Crossland in press). Here I simply want to establish the way in which these decorations operate, through contiguity. As Bloch points out, the designs of the carvings do not encode any meaning, but rather they evoke a wider set of associations through practice. They draw attention to the wooden parts of the house, particularly the central house posts, and they index the men who sculpted the designs (see Kus & Raharijaona Reference Kus and Raharijaona2000 for the general significance of house posts in highland Madagascar). Most importantly, the carvings celebrate the ‘lasting qualities’ of the dense tropical hardwood (Bloch Reference Bloch and Kuper1992, 140). Here the time and attention spent carving is itself an index of the wood’s central role in Zafimaniry life, pointing to its significance as a metaphorical resource to think about human maturation, ageing and ancestrality. In this way the carved designs connect with wider symbolic and metaphorical fields, despite not ‘representing’ anything themselves. Lakoff and Johnson suggest that this is central to the work of symbolic metonymies, which forge links between the quotidian daily round and wider metaphorical domains. These can support metaphysical claims, religious, spiritual or cosmological, and offer an experiential grounding for more esoteric or intangible concepts (Lakoff & Johnson Reference Lakoff and Johnson2003, 41). A focus on material metonymy opens up the question of how associations shift from simple resonances to more abstract concepts; from affects to arguments (while always retaining the affective)—a question that Ian Hodder has positioned as fundamental to understanding the entanglements of people and things (Reference Hodder2012, 119–37).

In tracking material metonymy as a form of indexicality I am most interested in part–whole relations (Knappett’s ‘factorality’) and how these may be traced archaeologically by tracking how elements of craft move within and across different materials (cf. Sennett Reference Sennett2008). This kind of sideways meaning-making seems important in craft work. We often see it archaeologically, in the mortice-and-tenon construction of Stonehenge for example, or the many examples of skeuomorphism, where some materials are imprinted with the characteristics of others (e.g. Ortman Reference Ortman2000). Knappett has written thoughtfully on the semeiotics of skeuomorphs in Minoan Crete (Reference Knappett2002). He describes how some ceramic vessels were made to look like metal, with the addition of ‘rivets’, while others were made to look like baskets by pressing the clay into a basketry mould and taking an imprint (Reference Knappett2002, 109). Knappett notes that a relationship of similarity (an iconic relation) characterizes the first form, while the second is related indexically, in the sense that it is partially caused by, and so connected to, the basket. He argues that therefore indexicality is only involved in some kinds of skeuomorph, such as the basket-moulded ceramics. This seems to be a clear-cut case. However, if we think about the kind of indexical part–whole relationships that often inform metonymy, we can see them at play in both forms. Informing the relationships of similarity in the metal skeuomorph is a kind of synecdochic relationship. In the same way that digging can evoke a whole realm of archaeology, the clay rivets evoke a metal vessel. They do not need to look particularly metallic as long as they are the same size and shape and index those placed in similar locations on the metal prototype. Indexicality therefore plays an important part in constituting likeness or iconicity (cf. Ahmed Reference Ahmed2004, 119). Icons and indexes are notoriously difficult to separate out semeiotically (Peirce Reference Peirce1931, 2.247), although one relationship may seem to predominate over the other at different times. The combination is particularly powerful in creating mimetic associations between copy and object (Taussig Reference Taussig1993). Likeness may be closely specified, or more loosely gestured toward, and it is these latter cases that interest me here in terms of their metonymic operations.

Below I consider how elements of craft materials have travelled from one craft domain to another in the history of pottery making and use in the highlands of Madagascar, asking what gets bundled together with the decorative work of craft as it travels? What kind of affective associations are summoned and extended, and what kinds of claims are expressed and built upon the sticky feelings that get caught in things? When similar elements move between different media, connecting different bodies and classes of material culture—from ceramic forms to basketry, to wooden sculpture, to metal objects or to the bodies of humans and animals—disparate domains are associated, and this association allows a politics to be articulated that may be emergent or not (yet) sayable. As archaeologists attentive to changes over time in material culture, we can trace ‘the “absent presence” of historicity’ and explore how it may be expressed through persistent metonymic connections that appear and are sustained at different periods. I am particularly interested in how the qualities of objects can act as signs that evoke or point to the feelings involved in political discourse and action, drawing attention to some objects or domains and eliding others. How might this allow us to consider the kind of implicit affective arguments that circulate within and around political formations? I end by considering counter-affects, and the ways in which material culture might affirm values and feelings that go against the political grain and how in turn this might bring invisible actors into view.

Like the lip of a vessel

At the start of the nineteenth century, ceramic forms in the highlands of Madagascar were characterized by their lustrous graphite coating and dark reduced interior fabric (Fig. 1). The court of the highland Merina King Radama I certainly used European ceramics, brought to the island as part of the ever-increasing flow of foreign missionaries and traders. But the most important traditional vessel was the loviamanga or ‘blue dish’, a tall, pedestalled bowl, used for consuming rice and its accompaniments (Rasamuel Reference Rasamuel2007, 74, 193–5). Susan Kus has explored the meanings of blue in relation to the king’s ancestral home of Ambohimanga (the Blue Height), noting the association of blue with excellence and beauty (Reference Kus, Ingersoll and Bronitsky1987, 358; also see Hebert Reference Hebert, Allibert and Rajaonarimanana2000). This was a time of great change in the highlands and elsewhere as the highland kingdom started to enter into the European political and economic sphere, with a treaty confirmed and signed by Radama in 1820 to prohibit the external trade in enslaved people and to receive a payment from Britain to offset the loss of income (Campbell Reference Campbell1981). Radama sent a group of young men to Britain and Mauritius under the aegis of the London Missionary Society to be trained in the production of textiles, gunpowder and other industrial techniques (Ayache Reference Ayache1994, 73–5; Campbell Reference Campbell2022). Artisans also came to Madagascar to train people in the same techniques, sponsored by the LMS (Cameron Reference Cameron1874). Although the external trade in slaves was prohibited, the domestic trade in people continued unabated. People were also stolen and sold abroad by raiders from regions that remained outside the control of the highland kingdom (Campbell Reference Campbell2005; Larson Reference Larson2000).

Figure 1. Loviamanga: grey graphite-covered rice dish. Cf. Rasamuel Reference Rasamuel2007, 194.

The elegant blue-grey vessels were the traditional vessel used for eating rice, going back before the time of Radama’s father, the state founder, Andrianampoinimerina. The consumption of rice was associated with the proper flow of ancestral blessing and the continued fertility and wealth of the kingdom (Berg Reference Berg1981; Bloch Reference Bloch, Rowlands and Friedman1977). Famously, Andrianampoinimerina spoke to his son on his deathbed, telling him that now that the kingdom had been gathered together, the sea was the limit of his rice paddies.Footnote ⁴ In this way he directed his son to conquer and control the entire island. The social and political worlds of the highlands revolved around this crop. Hidden behind the great respect given to rice and the land and households that produced it was the profound and enduring alienation of enslaved people from their own rice lands and sources of blessing. The sale of the enslaved into the international market was the primary source of wealth for Andrianampoinimerina and his advisors, as explored by Pier Larson (Reference Larson2000).

Nineteenth-century Madagascar was a society where the figurative play of language was highly valued, especially in political speech-making; it remains so today (Domenichini-Ramiaramanana Reference Domenichini-Ramiaramanana1983; Haring Reference Haring1992; Paulhan Reference Paulhan1913). In one oral history recorded in the 1860s, it was remembered that the subjects of the kingdom told Radama that they kept the words of his father, like money or pearls. His words were an inheritance which had to be nurtured and increased (quoted in Larson Reference Larson2000, 211). The inheritance of the ancestors was fundamental to how the legitimacy of rule was imagined, and this could be demonstrated both through recounting of history and through the inherited material conditions of the kingdom (see too Kus Reference Kus, Ingersoll and Bronitsky1987; Reference Kus, Miller, Rowlands and Tilley1989; Kus & Raharijaona Reference Kus, Raharijaona, Inomata and Coben2006). As more missionaries and Europeans arrived, they documented the rich poetic language of politics they encountered. There was a particular focus on collecting proverbs (ohabolana) from across the highlands (e.g. Cousins & Parrett Reference Cousins and Parrett1885; Houlder Reference Houlder1894). These were pithy philosophical nuggets, drawn from life and familiar to all, deployed creatively in poetry and speech-making to evoke different political possibilities or prohibitions, as they still are today. Like the ancient pots brought out for the Royal Bath festival, proverbs were, and continue to be, redeployed and recycled in political speech to bring the past and the present into close relation (Jackson Reference Jackson2013, 126–30). Many proverbs deal with material culture and particularly pottery. One in particular offers an entry point for thinking about politics through craft materials: ‘Like the lip of a vessel so we form a single circle’.Footnote ⁵ This was glossed by Jesuit missionary Paul de Veyrières (Reference Veyrières1967) as speaking of peace and harmony in the kingdom or in the family. It remains current in today’s speechmaking practices and has some antiquity. Oral histories of the 1860s remembered Radama’s father deploying the phrase in his speeches. In a discourse on capital crimes and obedience to the laws of the kingdom, he was said to have made the claim that his subjects formed a single circular unity like the rim of a vessel—there was no distinction between them; instead all were united in the same way as the front and back of a traditional piece of weaving, each side indistinguishable from the other (Callet Reference Callet1908, 757; also see Kus Reference Kus, Ingersoll and Bronitsky1987, 357).Footnote ⁶

Here we can see that communal peace was predicated on working together in harmony with others and with the laws handed down from the ancestors. This also meant that subjects should act together under the King’s guidance, not standing out, and not disobeying. When Radama’s father described how his kingdom had been assembled, his remembered speech uses the word vory for this gathering, which had a secondary meaning of round or circular. So in the oral histories of the nineteenth century, the late eighteenth-century state was imagined in terms of an encircled and bounded gathering, bringing to mind again these ceramic rims. This is a metaphor, of course. The community is the vessel; the vessel stands for the community, whether household or kingdom. But what is the metonymic ground for this? How was this metaphor embedded in highlanders’ experience of the world and of pottery? We can see that the circular rim of a vessel was a feature that needs to stay unbroken and continuous for the vessel to function as it ought. So there is a kind of functional metonymy there—a part–whole relation in the way that the lip stands for the vessel as a whole—that offers the ground for the metaphor. We might also think about the ways in which vessels summon community and household, particularly in dining and feasting practices—not only are the living drawn together around the rim of a pot, but the ancestors too are present, not least through the sacred rice that was served in these striking grey-blue vessels.

The focus on the lip of the vessel in this proverb also provides a locus that we can consider archaeologically. In the pottery of highland Madagascar, careful attention was often paid to vessel mouths, but this was expressed differently through successive highland pottery traditions. This attentiveness, and the quality of that attention, suggest that practices of decorating and coating vessels might index wider concerns in earlier periods as well. This is not to argue that the political ideals of the nineteenth or twentieth centuries can be transposed into the deeper past; rather that the attentiveness paid to the rim of ceramic vessels offers a site of elaboration that we can try to connect with other contemporaraneous concerns and their changes over time. In this way we may also start to see the ‘absent presence’ of historicity in the vessels so carefully preserved in the Queen’s Palace. Whenever this proverb emerged—whether it was a nineteenth-century coinage or had more time depth, I would like to suggest that the treatment of pottery vessels offers a strand of evidence to explore how community was affectively constituted (cf. Léon Reference Léon, Chami, Pwiti and Radimilahy2003).

The first ceramics that are found in Imerina date to the twelfth to fourteenth centuries and consist mostly of simple globular jars and carinated bowls, some of which had pedestal bases. The clay fabric contains quartz sand inclusions, but the ceramics have a smooth compact surface, sometimes with traces of graphite clinging to it (Wright Reference Wright2007, 41–5; Wright & Kus Reference Wright, Kus and Kent1979). These earliest pots were often decorated with incisions and triangle impressions filled with white clay to create a contrast (perhaps skeuomorphic of the light areas on carved gourd containers). Similar motifs were probably also present in organic materials such as carved wood and basketry, which unfortunately are now lost to us. What is noticeable about these highland ceramics is that, as new forms of social distinction emerged in the fifteenth to early sixteenth centuries, the impressed and incised decoration became more elaborate, as did the variety of vessel forms (Wright Reference Wright2007, 42–7). This was a time when people moved up to higher hilltops and created larger bounded settlements. We see great efforts invested in the construction of ditches around villages throughout the highlands (Crossland Reference Crossland2001; Raharijaona Reference Raharijaona1988; Wright Reference Wright2007). Arguably we can already see a concern with boundedness and with gathering at this time, with the physical containment of the village within a delineated space.

These small hilltop sites were connected with the Indian Ocean world from the earliest settlements in the highlands, as evidenced by imported trade beads and ceramicsFootnote ⁷ These include green glazed (‘celadon’) bowls from southern China, with a flattened ledge-lip and flowing decoration. We find traces of these alongside plentiful locally produced ceramic vessels that copy the characteristic form of the imported ledge-lipped bowls. Retaining the shape and dimensions of the Chinese green-ware bowls, they are made in a local idiom, of low-fired earthenware, covered with graphite and alive with impressed decoration highlighted with white kaolin (Fig. 2). I sense a delight in making here, in the metonymic seizing of an exotic, scarce, and presumably highly valued ceramic form and a remaking of it for local consumption. In this bridging of local and imported pottery, the earthenware ceramic form effects a kind of attachment between the imported green ware and local domains of pottery manufacture and food practices. On the locally produced ledge-lip bowls, the incised and impressed decoration travels around the ledge of the rim, and sometimes also folds over from the interior to cover the exterior surface, asserting a continuity with other local bowl forms. These commonly had complex repeating and multi-part designs on their exteriors. There is an exuberance to the ceramic decoration at this period, especially between the fifteenth and sixteenth centuries, a feeling of experimentation and extemporization, a space for the agency and imagination of the makers (Fig. 3). Designs become more elaborate, but also stabilize around a repertoire of patterns toward the end of this period (Wright Reference Wright2007, 48).

Figure 2. Imported and locally made ledge-lip bowl sherds.

Figure 3. Impressed decoration of the fifteenth century.

Bowls may have had the most elaborate decoration, but cooking vessels and jars for storage were also decorated in a band on the shoulder, just below the rim. The time spent in the decoration of all these ceramics suggests that whoever was making them was invested in the social practices around food production, storage and consumption. These are unlikely to be objects that were made by craft specialists, even though some examples show a great deal of skill, both in the making and the decoration. There is plenty of variation in form and decoration, even within a single vessel. Vessels were probably fired in the open or in shallow pits, and the clay seems to have been sourced locally, without too much concern about the particular inclusions it contained (cf. Rasamuel Reference Rasamuel1982, 135–6; Reference Rasamuel2007). These pots were probably being made within households, quite likely by women. Certainly, this was the case for the historically documented pottery that was made in highland Madagascar several centuries later (e.g. Ellis Reference Ellis1838, 1.320).

Impressions fall out of fashion

At some sites there is a transformation in ceramic forms around the same time, in the fifteenth to sixteenth centuries (Wright Reference Wright2007, 50–51). Impressed and incised decoration is eschewed for smooth, highly burnished ceramics, especially pedestal bowls covered in graphite and well fired, with a surface that takes on the smooth and lustrous sheen of metal (as shown in Figure 1). These ceramics are no less well made than earlier forms—if anything they are technically more proficient—the high-quality graphite burnish is dependent on firing in a controlled environment where oxygen is not available (Kreiter et al. Reference Kreiter, Bartus-Szöllősi, Bajnóczi, Havancsák, Tóth, Szakmány, Alberti and Sabatini2013, 176; Yiouni Reference Yiouni2000, 209). This style of surface treatment comes to dominate, becoming common during the sixteenth to seventeenth centuries and also being used for the surface of jars as well as bowls. It persists as a decorative mode until the loviamanga we know from the nineteenth century (Wright Reference Wright2007, 52–8). It seems no coincidence that this metallic graphite burnishing starts to take on more importance alongside an increasing presence of imported guns and silver coins. Both were intimately involved in the work of capturing and enslaving people and building political power. The first European contact with Madagascar began in the early sixteenth century after Portuguese navigators spotted the island in 1500 and named it São Lourenço. From that point onwards the south and west of the island saw numerous shipwrecks and trading visits, as well as attempts at exploration and colonization by both the Portuguese and the French (Larson Reference Larson2007; Parker Pearson Reference Parker Pearson1997; Wright & Rakotoarisoa Reference Wright and Rakotoarisoa1998). Guns began to circulate in greater numbers and more widely across the island. We should also remember that spears were a highly prized component of male identities at this period. Even by the eighteenth century spears remained in common use in raids to seize people from other villages, and were viewed as more reliable than guns (Berg Reference Berg1985, 265; e.g. Drury Reference Drury1890, 59, 183–4). Muskets also came to play an important role in political ceremonies of allegiance where male subjects swore a blood oath of loyalty to their leader upon them (Berg Reference Berg1985, 266).

The serving and eating bowls take on the associations of metals through their shine and their colour; their surfaces assert a visible metonymic connection, which travels from one highly prized item to another. We can think about the silvery-grey colour of a slipped and burnished ceramic vessel in terms of qualia—the qualities that are distinguished in sense perception. Although such qualities may seem self-evident, one must of course learn to recognize them, to have a feel for them; a process that takes place under specific historical and political circumstances (see Chumley & Harkness Reference Chumley and Harkness2013). How the recognized boundaries of a colour—its hue, chroma and value—are defined depends on the context, on how and where it is expressed, and who recognizes it as ‘grey’ or ‘blue’ (see Jones & MacGregor Reference Jones and MacGregor2002). Recognition of the surface qualities of graphited vessels also encompasses the felt-understanding of this greyish shiny quality in other contexts: qualia come with feelings attached. In its instantiation as part of a pottery vessel, the surface takes on some of the qualia of the vessel and its uses. Furthermore, the quality becomes known and understood through its deployment at other material and linguistic sites. In so far as this quality provokes a feeling for burnished metallic-seeming surfaces and for other associated meanings, it acts as a qualisign, a quality that is meaningful, creating an interpretive field that is filled with potential rather than tied to something more concrete (such as a proposition or an argument). We can trace the flows of affect through the interplay of qualia, the ways in which they act as signs of feeling, and the habitual and repeated practices of the semeiotic community that inform how they are perceived and reproduced. As particular qualia are redeployed in new object-contexts the associated affective load can travel and morph alongside them. This may foster feelings of recognition and perhaps acknowledgement of the efficacy (or not) of a particular political formation. I suggest that in highland Madagascar guns, spears, coins and pottery became associated metonymically, and that in the process the violence extended through raiding and fighting was linked to the positive associations of home and its ancestral denizens. This suggests a political affect that connected the work of raiding with the security and reproduction of the household and village and its associated agricultural land.

Among these highly graphited vessels, the earliest dishes or plates were sometimes impressed with circular or triangular motifs along the lip, bordered by parallel incised lines (Rasamuel Reference Rasamuel2007, 161). Over time the graphite coverage became more complete and uninterrupted. This worked to merge the lip of the rim with the rest of the vessel in one coherent unity. There is less concern with emphasizing boundedness here, and more of a feel for coherence and consistency. By the seventeenth century, ceramics in Imerina were no longer the site of decorative experimentation in impressed surface treatment. Effort was instead directed towards the change in firing technology, to allow more complete and reliable coverage of the vessel’s surface by the graphite. The complete graphite coverage demanded extended time spent burnishing and polishing to a shiny finish. There may still have been a certain enjoyment in their making, but this seems to be subordinated to the more unified and consistent expectation of a plain reflective surface. Although incised and impressed forms of decoration largely disappear, they do remain in attenuated forms on a few bowl rims and pedestal feet, and below the rims of cooking and storage vessels (Wright Reference Wright2007, 52–3), perhaps more associated with the domain of women and food preparation. There seems to be an affective shift here, a changing set of domains and concerns that are evoked through these vessels—a focus on consumption contexts tied to a validation of metals and masculinity and their associated practices, and away from the impressed decorative motifs that were previously found on pottery and perhaps elsewhere in house and village. It is perhaps also significant that the sole decorative motif that persists with any frequency in Imerina through until the early eighteenth century is a single line encircling the vessel, and incised just below the rim on cooking and storage jars (Wright Reference Wright2007, 54). This is located at the point where the rim meets the jar shoulder, where two strips of clay meet in the pot’s manufacture. This is a common place of breakage, so it marks and holds a site of weakness in the vessel. There is a divergence here in how different vessels are marked—those pots for household use carry slight traces—an ‘absent presence’—of the older incised decorative practices, and they contrast with those made for those more masculinized spaces of visible consumption and sharing, which also evoke more threatening absent presences in the form of guns and spears. What seems to be emerging is a feeling for the importance of unity and community in the context of increasing threats to it. This would only intensify over time as raids became more frequent and more closely involved with the international trade in enslaved people.

Changes in affective value

By the late eighteenth and probably into the early nineteenth century, bowls are still highly graphited, but storage and cooking jars are less likely to have graphite on them (Wright Reference Wright2007, 56–8). We see much more of a division between the burnished vessels used for consumption and the plain unadorned vessels made for food preparation. The graphite surface treatment is also often less adherent, rubbing off more easily (Wright Reference Wright2007, 56). Towards the end of this period, vessels appear that emulate imported cast-iron pots, with raised ridges running along the circumference and at right angles. These changes suggest there is a shift and maybe diminution of the ways that women’s work around ceramics was valued as part of the public sphere at this time. Women’s efforts were directed elsewhere, including towards corvée labour for the newly formed state. We know by the nineteenth century women were weaving the beautiful textiles widely used in the highlands, another high-status good that could be publicly displayed (Peers Reference Peers, Kusimba, Odland and Bronson2004). Those made of silk were particularly important in the context of politics and exchange (Fee Reference Fee, Kreamer and Fee2002). Unfortunately, we have little information, archaeological or otherwise, on textile production from earlier periods, and so cannot track changes in weaving, which surely took place alongside those in ceramics.

The burnished ceramics of the seventeenth to eighteenth centuries are associated with a period of struggle between rival polities, and with the emergence of an increasingly masculinized sovereignty, ultimately ending in the formation of the powerful and expansionist kingdom founded at the end of the eighteenth century by Radama’s father (Larson Reference Larson2000). By the late eighteenth century the image of circular and undifferentiated unity documented in the 1860s had probably indeed emerged, linked to the development of new forms of highly stratified political power—just as the oral histories recount. This created a new political identity for those who belonged to the highland state, grounded in genealogical claims to power and land (Larson Reference Larson1996). The importance of unbroken circularity was reiterated in the uncut silver coins (vola tsy vaky) that were given to the king as tribute (Bloch Reference Bloch, Rowlands and Friedman1977, 19; Kus & Raharijaona Reference Kus, Raharijaona, Bauer and Johansen2011), and also in the massive circular stone gates that were rolled across the entrances to settlements associated with the emergence of the kingdom under Andrianampoinimerina (Gabler Reference Gabler and Wright2007). This was power that was imagined in an idiom of the household (Kus & Raharijaona 2000; Reference Kus, Raharijaona and Richard2015), in which women could also play a role, often through marriage, and sometimes as ruler (Raharijaona & Kus Reference Raharijaona and Kus2010), even if the oral histories tend to emphasize the actions of royal men. Yet alongside the emphasis on unity and communal harmony, the oral histories are largely silent on the violence that was at the heart of the state-building process (Larson Reference Larson2000). The state was built in part on the wealth accumulated from the trade in enslaved people, and also through corvée labour, whether military service, craft production or agricultural work (Campbell Reference Campbell1988), putting pressure on the ordinary tasks of the household. This seems to be reflected in the changes in ceramics. By the nineteenth century, locally made ceramics were used mostly for cooking and storage, as at the Royal Bath festival. Less attention was paid to their decoration, and graphite became less important as a form of surface treatment. Yet a small amount continued to be applied by potters to the lip of cooking pots (Wright Reference Wright2007, 58), a metonymic indexing of the rim, in its circular and unbroken entirety.

A feeling for red

Although the trade in external slavery was banned by Radama in the early nineteenth century, internal slavery for domestic purposes continued. And the external trade continued unabated in regions that were not within the control of the highland kingdom. Corvée demands also increased. This led to a great deal of unrest and political instability in the latter part of the century as the state, its army and administrators continued to build wealth through violent practices of raiding and extortion. In frontier areas, such as the highland region of the Vakinankaratra to the south, there was at times almost complete breakdown. The administrators and military men who were put in place to control the violence were among its worst perpetrators, building wealth and prestige through raiding and extortion (Crossland Reference Crossland2014, 181–227; Esoavelomandroso Reference Esoavelomandroso1989).

Sometime in the nineteenth century, a new ceramic form appeared in the Vakinankaratra frontier region and further south, especially at sites closely associated with vassal rule. Small, fine, red-slipped bowls with a low ring or pedestal base began to be made and used; by the late nineteenth to early twentieth century they were in common use. Examples were collected by Sibree in 1892, and by Ralph Linton in 1925-6, and are now held at the British Museum and the Field Museum respectively (Fig. 4). The bowls’ lustrous red slip was coloured with the iron oxide that stains much of the lateritic soil of the highlands—and it marks a major shift in ceramic decorative treatment. Prior to this moment it is rare to see red-slipped ceramics anywhere in the highlands outside of Antsihanaka, far to the north. This change from the dark gun-metal grey of traditional vessels that had persisted over centuries to a startling red form is dramatic and seems to respond to the political moment. This is a shift that is unlikely to have been discursively negotiated. Instead, we have to think about the political and affective stickiness of the colour red. What did the colour evoke in nineteenth-century highland Madagascar? What feelings did it bring up for the makers and users of these vessels?

Figure 4. Red-slipped bowl of type collected by Sibree, held at the British Museum.

Social unrest and disorder intensified at the end of the nineteenth century as mission sites came under attack, violent raids involving murders, abductions and cattle theft were regularly made on towns in border regions, and corruption was rampant (Campbell Reference Campbell1991). In 1883 the French attacked Madagascar from the north and were bought off by the Crown before again attacking in 1894, this time succeeding in capturing the island and establishing Madagascar as a French protectorate in 1896. In response to the French attack and the local policies of the highland state, a popular rebellion emerged to the west of Imerina and spread quickly through the highlands (Ellis Reference Ellis1985). Large fires were set and missions and military garrisons attacked. The rebels called themselves the menalamba, and although they opposed the French invasion of Madagascar they also fought against the oppressive rule of the state and against European and Christian influence (Campbell Reference Campbell1988). They called back to older traditions of royal governance that conformed with ancestral precepts. These rebels became known for their muddy red wraps or lamba, stained with the earth of the highlands (Kus & Raharijaona Reference Kus, Raharijaona and Richard2015). Lamba wraps made of cotton or silk were commonly worn by people in the highlands at the time, and were used both for the living and the dead (Mack Reference Mack1989; Peers Reference Peers, Kusimba, Odland and Bronson2004). Stephen Ellis, who studied this revolt, translated menalamba as ‘red shawls’ or red lamba. He suggested that in emphasizing the colour red the rebels drew associations with royalty, and asserted the historical authority of the sovereigns against that of foreigners. The silk lamba used as shrouds for the dead were called lambamena, so the revolt also inverted the traditional name for burial shrouds and signaled its ancestral legitimacy. Although Ellis glossed the menalamba as ‘red shawls’, it is perhaps more accurately translated as lamba-red, or the red of lamba—this brings out another dimension, in that it is the colour that is foregrounded, rather than the woven wraps themselves. Red had powerful associations for Malagasy highlanders. Writing at the end of the nineteenth century, Norwegian missionaries to the Vakinankaratra described red as the colour of rule, a colour which conjured a sense of duration and perpetuity (Vig [1892] 2001, 179; [1902] Reference Vig1977, 14–15; Dahle quoted in Sibree Reference Sibree1896, 285). The symbolism of colour may be unstable, of course, its meaning shifting according to context, as Victor Turner showed with his analysis of Ndembu colour symbolism (Reference Turner and Banton1966). Yet, if we think about how perceptions of colour were grounded in experiential relationships, it is possible to trace relatively stable metonymic linkages that offered the ground for this symbolism of rule and its permanence, the feeling for red which was associated with sovereignty.

The colour red was widely used in royal ceremonies such as the Royal Bath, and sumptuary laws controlled its use. Oral histories recorded in the 1860s recalled that when the frontier region was first conquered the state’s representatives in the Vakinankaratra were given a red hat and red lamba. This gift demonstrated their legitimate rule under Andrianampoinimerina (Callet Reference Callet1908, 990–1002). In 1870 Sibree described the opening of a new palace in Antananarivo and described how ‘older men, heads of tribes and chiefs of villages wore the red lamba, used at marriages, at the new year’s feast, and at other festivities’ (Sibree Reference Sibree1870, 352–3). He also recorded how, when a new gateway to the royal courtyard was excavated at Antananarivo a few years prior to writing, a foundation deposit containing ‘the remains of one of the former queen’s fighting-bulls [was] discovered, carefully wrapped in a red lamba’ (Sibree Reference Sibree1881, 46).

The red of the bowls points to this complex history, acting as a qualisign that evoked by metonymic connection the red of the sovereigns, of rule, and of the dead. But it also suggested the ochreous red of the earth from which the bowls were made, and by extension the rice land inherited from the ancestors. This connected the everyday practices of the household with the great history and ceremonies of state and with localized claims to place. To make and use these bowls was to gesture toward things that were difficult to speak of in a context that was fraught with danger and insecurity. In doing so the bowls created the metonymic ground for an assertion of continuity and legitimate local rule in the face of multiple threats. The ceramics appear at a time of breakdown, when the established political organization was threatened and at risk of collapse. Paradoxically, the innovation of the red pottery worked to confirm and assert the legitimacy of local elites, building upon established associations with royalty. Just as with the menalamba revolt, this was a call back to an older time, to ancestors and sovereigns past and a refusal of the contemporary chaos. We also seem to be seeing a popular intervention by women into politics here, one that is expressed in a cherished domestic space of consumption which also provided many of the metaphors on which the kingdom was founded. This intervention most likely operated affectively and non-discursively, evoking the pleasure associated with crafting pottery and making and serving food, the role of these practices in sustaining the family and community, and bringing these domestic tasks into relation to a claim for a geographically grounded continuity of ancestry and land at a period of chaos in political life. This intervention not only resonated with the politics of the menalamba rebellion, but can be juxtaposed against the successive queens who acted as head of state together with their prime minister consorts, ruling conjointly for much of the nineteenth century (Rajaonah Reference Rajaonah2020).Footnote ⁸