The Other Shoe: Fragmentation in the Post-Medieval Home

Deposits are not always recovered whole; many are found broken and damaged. The obvious explanation is that such objects were accidentally broken; however, some have been interpreted as having been deliberately damaged by their depositors, a practice termed ‘frag-mentation’. Objects are broken into parts and deposited incomplete, often in ways that make their missing parts starkly evident. Thus many fragmented deposits denote synecdoche. It is the position of this paper that the absent (part) is just as integral to an understanding of the whole as the present (part) is, and this notion is explored by focusing on the post-medieval concealed shoe: an item of footwear that was fragmented by being deposited within the fabric of a building without its counterpart, for reasons unbeknownst to us. Drawing on a sample of 100 examples, this paper questions why such shoes were deposited as singles (the present parts), what became of the ‘other shoe’ (the absent part), and how such consideration aids our understanding of this enigmatic custom.


Introduction
Archaeological deposits are not always recovered whole; many are found broken and damaged, with missing parts and incomplete forms. The two most obvious explanations are that such objects were accidentally broken either before deposition or during that (often significant) period of time between deposition and recovery (e.g. Martin & Meskell 2012, 405). However, another explanation is that some deposits were deliberately broken by their depositors (cf. Oates 1966, 150;Talalay 1987;1993), a practice termed 'fragmentation' within archaeological theory (Brück 2006;Chapman 2000). Objects are broken into parts and deposited incomplete; not subtly or obscurely, but often in a way that makes their missing parts starkly evident. Thus many fragmented deposits denote synecdoche: as Chapman writes, 'the (present) parts clearly signify the (absent) whole ' (2000, 104).
Examples of such synecdoche are numerous, with Chapman noting that there are 'thousands of cases of the "missing fragment" ' (2000, 54). These include axes from Late Bronze Age hoards, deliberately cut into two halves before deposition, the mouth and the blade rarely appearing in the same hoard (Brück 2006, 310). In some cases only a small fragment is found, with the remainder of the object notably absent; the large-scale excavation of Polgár, a Late Neolithic flat site, for example, recovered a small sherd of a ceramic bowl with no matching fragments found on site (Chapman 2000, 64). In other cases, whole hoards consist of unmatched fragments; for instance, of the 838 bronze fragments discovered at the Bronze Age site of Polešovice, Moravia, only three fragments were found to match (Hansen 2013, 180).
The present part and the absent whole are obviously central to analyses of this surprisingly widespread practice, but it is often easy to neglect the absent part: that component of an object which is removed and not recovered. But as Meskell observes, 'just because something is not on public view … does not necessarily mean that the object is not working ' (2008, 237). Meskell here is referring to Mills' (2008) work on dedicatory offerings, such as strings of beads secreted away in niches of the buildings at Chaco Canyon, a prehistoric complex in the southwest USA. Mills argues that such rituals of concealment are central to processes of memory making: 'Although out of sight, sometimes permanently,' she writes, 'the location of those objects may be remembered for long periods of time ' (2008, 82). Likewise, when a deposit is fragmented and its fragments distributed, the Ceri Houlbrook depositor is aware of, and remembers, their respective locations-even though we, as archaeologists, are not.
It is the position of this paper that the absent part is just as integral to an understanding of the whole as the present part is, and will explore this notion by focusing on a specific example of fragmentation: the concealed shoe.

Introducing the concealed shoe
The concealed shoe is-as one might imagine-a shoe that has been concealed within the fabric of a building, often a single shoe, found without its counterpart. This is not a paper about the concealed shoe, which is the focus of numerous other studies (cf. Evans 2010;Evans et al. 2016;Hoggard 2004;Houlbrook 2013;Manning 2012;Merrifield 1987;Swann 1996;. Instead, the focus of this paper is on the other half of the pair: the shoe that is not discovered. However, in order to contemplate this, a (necessarily brief) summary of the custom of the concealed shoe is required.
The concealed shoe eluded academic attention for some time. It was only in the 1960s that June Swann, former Keeper of the Boot and Shoe Collection of Northampton Museum, first began to publish about the custom, having noticed a recurring pattern in the finds being donated to her department (Swann 1969). A range of primarily eighteenth-and nineteenth-century footwear was brought to her attention, having been discovered in unusual locations within buildings: in the roof space, fireplace, chimney breast, walls, doorways and foundations (Swann 2016, 123). Discounting accidental loss for most of their locations, Swann began to gather as much data as possible on this unfamiliar, previously unstudied phenomenon (Swann 2016, 119), instigating the Index of Concealed Shoes (hereafter the Index).
From its moment of conception, the Index grew significantly: from 129 in 1969 to 700 by 1986700 by (Merrifield 1987; to 1550 by 1996; currently standing at close to 2000. These shoes represent a wide variety, belonging to men, women and children, both workingand upper-class styles. Most of them have been dated to the eighteenth and nineteenth centuries, although some pre-date the 1700s and a few post-date 1900. The vast majority of them were discovered in England and Wales, although the custom is not limited to Britain, with a number having been discovered as far afield as Australia (Evans 2010) and North America (Manning 2012).
For the purposes of this paper, a sample of 100 concealed shoes and caches have been selected, detailed in Table 1. These examples have been variously sourced from the Index, from academic publications, media publications, and personal correspondences. While the individual cases were selected at random, the sample is intended to be representative of the custom in Britain, with cases distributed throughout England (both northern and southern counties) and Wales, from a variety of dates and building types. The statistics given throughout this paper are based on this sample and are intended to approximately exemplify rather than accurately apply to all concealed shoes, the vast majority of which undoubtedly remain unrecorded.
The purpose of the concealed shoe still remains a mystery to us, simply because no contemporaneous written record has been identified describing the practice and elucidating why these shoes were concealed. 1 Theories certainly abound, and while the author is resistant to a one-size-fits-all approach (forgive the pun) in assigning a single purpose for all concealed shoes (see below), for the purpose of this paper the most popular theory is considered: that concealed shoes were a category of apotropaic device, employed to protect a household against malevolent threats.
The strongest evidence to support this is the locations of the shoes: in chimneys, hearths, walls, roof spaces and under floorboards, all locations that were viewed in the early modern period (defined in this paper as stretching from the sixteenth century to the late eighteenth) as being particularly vulnerable to the myriad supernatural forces, from demons and witches to ghosts and fairies, which threatened to infiltrate the home (Lloyd et al. 2001, 57;Reay 1985, 116). Such liminal, assailable domestic spaces have yielded numerous other probable apotropaic devices of the period: dried cats, horse skulls, timber markings, witch-bottles and various other garments. 2 The shoe, however, appears to have been the most commonly concealed object in England (Hoggard 2004, 178), and there are a number of theories regarding its popularity. Firstly, the shoe's material; in popular belief, fairies, demons and spirits did not like the smell of leather (MacCulloch 1910, 202;Radford & Radford 1948, 306). Secondly, the shoe's shape: bowl-or vessel-like, explaining how John Schorne, a thirteenth-century parish priest from Buckinghamshire and one of England's unofficial saints, was said to have captured a demon within a boot (Merrifield 1987, 135), subsequently leading to the shoe being perceived as a form of 'spirit trap' in popular belief (Hoggard 2004, 179). Thirdly, the shoe's association with fertility and luck (Houlbrook 2013, 16); and fourthly, its close association with its owner, often acting metaphorically to represent past and present wearers (van Driel-Murray 1999; White 2009). Swann notes that the shoe is 'the only garment we wear which retains the shape, the personality, the essence of the wearer' (1996,56). By retaining the foot's shape-and smell-the shoe can stand as substitute for the wearer, a quality that can prove very useful in customs designed to protect a household from malevolent forces. Easton (1999), for example, theorizes that the concealed shoe may have acted as a form of 'lightning conductor' in diverting the malevolent supernatural threat from entering the house; the invading witch, demon, or spirit 'sees' the shoe, believes it to be a member of the household and attacks the shoe instead, subsequently becoming trapped inside.

Introducing the other shoe
The majority of concealed shoes are singles, discovered alone rather than in pairs. 3 One example of this is the single man's straight, buckle shoe, dated from the mid to late eighteenth century (Figure 1), found in the wall behind the wainscoting of the Combination Room in the Master's Lodge, St John's College, Cambridge (Newman 2016). This is a trend observed (albeit often only in passing) in most studies on the subject, with Swann (1996, 65) noting that only 11.3 per cent of concealed shoes are found in pairs. This appears to have extended beyond Britain, with Evans et al. (2016, 236) noting that the majority of concealed shoes in the USA are also singles. Of the sample of 100 instances of shoe concealment this paper is concerned with, only six comprised pairs of shoes (Table 1; Figure 2). Four of these were pairs concealed on their own, and examples include the man's nineteenth-century elastic-sided, hobnailed boots discovered in the roof of the vestry of the Savoy Chapel, London ( Figure 3) and the ankle boots found in the roof space of the Three Cocks Hotel in Brecon, Powys. The other two examples of pairs were in amongst caches, defined by Pollard (2008, 55) as 'single-event, multiple depositions'. These consist of multiple shoes, ranging from three (Tufton, Pembrokeshire) to possibly 100+ (Nant Gwynant, Gwynedd).   At least 83 per cent of the concealers represented by this sample chose to conceal specifically single shoes, either alone or in groups: a significant enough majority to suggest that it was the norm, rather than the exception, to separate shoes and deposit one without the other. It is the stance of this paper that such separation of shoes was deliberate rather than accidental, integral rather than incidental to the custom of concealment. It is also the stance of this paper that the unrecovered shoe (referred to hereafter as the 'other shoe') is just as central to the analysis of this practice as its concealed counterpart. After all, shoes come in pairs; a single shoe is one half of a whole. They neatly fit Chapman's observation (2000, 104) that the 'form of many of the artefact classes selected for fragmentation is so distinctive that the (present) parts clearly signify the (absent) whole'. The other shoe (absent part) is so conspicuous in its absence that the fragmented pair of shoes (absent whole) is starkly evident, engendering the question of why fragmentation was clearly central to this custom.
To consider this, a more pressing question must be asked: what became of this absent part, the other shoe? Writing of fragmentation within an archaeological context, Chapman (2000, 54) observes that the 'hardest task is to find and match the different parts of once integral artefacts', a difficulty also observed by Brück (2006, 310) and Hansen (2013, 180), and the concealed shoe proves no different. To date, no single concealed shoe has been reunited with its counterpart. A number of reasons may account for this: (1) the other shoe did not enter a ritual context; (2) it was ritually destroyed/discarded; (3) it was ritually concealed elsewhere. This paper will consider each possibility in turn, starting with the theory that only one of a pair of shoes entered a ritual context.
It is certainly plausible that, while one shoe was concealed, its counterpart was simply disposed of or stored elsewhere. It would probably not 266  continue to be used as footwear, not only because the other half of the pair had been put to ritual use, but because the vast majority (97.81 per cent, according to Swann 1996, 59) of recorded concealed shoes are heavily worn or badly damaged, no longer capable of fulfilling their roles as comfortable footwear (Houlbrook 2013, 107-8). Perhaps, then, the other shoe was simply disposed of, recycled for parts, or retained in storage; either way, it is unsurprising that the counterparts of concealed shoes have not been identified.
Concealed shoes, often discovered during building renovations, are only recorded because of their unusual find spots; their finders suspect some significance in their having been discovered up a chimney or within a wall, and subsequently report them to a specialist. Had the same shoe been discovered in a rubbish heap or at the bottom of a wardrobe, it probably would not have been recorded, and would likely have ended up being disregarded and discarded. There is no Index of Unconcealed Shoes, and therefore no resource for researchers to draw upon in any attempt to match concealed shoes with their unconcealed counterparts.
Material evidence, therefore, can neither prove nor disprove the theory that the other shoe was simply disregarded. However, if it could be proved, what would it signify? Why would the depositors choose to conceal one shoe and discard the other? If the concealed shoe was designed to act as 'lightningconductor' or 'spirit-trap', or if, conversely, it was meant to repel malevolent forces, then surely two shoes are better than one; surely concealing both doubles their efficacy? This, however, does not appear to have been the logic followed by the 83 per cent of concealers of this paper's sample who appear to have only concealed singles-what, then, was the logic they were following?

The other shoe retained
If one shoe was concealed and the other retained by its owner, then this may have constituted a form of contract, as suggested by Evans et al. (2016, 245). Carol van Driel-Murray has proposed this theory for the enigmatic shoe finds in the archaeological records of Roman contexts. Shoes, shoe-soles and shoe fragments have been discovered amidst myriad ritual deposits in, for example, wells at a Romano-British rural site near Uttoxeter, Staffordshire (Thomas & Thomas 2010); at Rothwell Haigh, Leeds (Cool & Richardson 2013); Coventina's Well, Hadrian's Wall (Allason-Jones 1996, 118); and at Chenies Manor, Hertfordshire (Swann 1996, 65). In many cases these shoe deposits were singles; van Driel-Murray (1999, 137) cites the specific examples of a shoe sole found tucked at the back of the wooden construction of a well at Venray (Netherlands) and a child's sandal sole found on the bottom of the well at the Roman villa of Dalton Parlours, Yorkshire. Drawing on these examples, she proposes 'That shoes form a pair invites their use in contractual situations, primarily as a pledge of mutual obligations' (van Driel-Murray 1999, 136). One shoe, she suggests, was deposited/offered as a pledge to the deities-in return for protection, luck, or healing perhaps-while the other shoe was retained by the supplicant; 'Symbolic of the contractual vow, the shoe becomes imbued with supernatural power and thereby becomes the earthly manifestation of divine protection' (van Driel-Murray 1999, 136).
It is possible, therefore, that the concealed shoe was being offered as a votum, defined by Derk (1995, 113) as 'a temporary contract between man and deity'. Who the supernatural recipient of the votum was in these contexts is unclear; perhaps the depositor of the shoe was hoping to supplicate a benign domestic spirit-or appease a malign one-with the aim of ensuring their safety, prosperity or health, and the safety of their home. The other shoe may have been retained to symbolize this contractual vow between the depositor and the supernatural recipient of its concealed counterpart. If this was the case, then the retained shoe is not incidental to the custom but central to it; its retention is part of the ritual.

The other shoe discarded
Another possibility is that the concealers of these shoes were endowing their deposit with what Gell (1998) would term 'cognitive stickiness', whereby apotropaic devices are designed to confound malevolent forces, such as witches, demons, and spirits. Conveniently, such malignant forces were believed to share a tendency towards obsessive compulsion, and so intricate, complex, or unfinished patterns could act as 'demonic fly-paper', to use Gell's phrase (1998, 84), luring evil spirits in, distracting and binding them, thus impeding their passage. Celtic knot-work patterns are just one example of this, but there are many more which constitute apotropaic 'knots' (Gordon 2013, 211). Fishing nets cast over doors would delay the entrance of a vampire, who would be compelled to count all of the net's knots (Trigg 1973, 153). For a similar reason, nets or stockings would be buried with corpses to prevent them from rising and grains of various varieties would be strewn across graves so that any revenants, so consumed with The Other Shoe counting them, would never leave the graveyard (Barber 2010, 49).
Another probable form of domestic apotropaic device which may have employed cognitive stickiness in its efficacy is that of the timber mark. Compass-drawn hexafoils, alternatively described as daisy-wheels, and 'merels'-type (series of squares and rectangles) markings adorned doorframes, windowframes, wooden beams and fireplaces in many postmedieval homes, possibly to confuse invading spirits. As Matthew Champion (2016, 18) hypothesizes, 'evil forces, when encountering a line, will be compelled to follow it, or become hopelessly confused-thereby trapping themselves within the symbol'.
Perhaps the single concealed shoe likewise exploited the obsessive compulsion of supernatural forces. As posited above, a single shoe is one half of a whole; it is, therefore, an unfinished pattern. Perhaps the malevolent threat would be lured and bound by the cognitive stickiness of one half of a pair-or perhaps, even more likely, they were confounded by it. It is, after all, not only creatures of folklore who feel the compulsion to complete unfinished patterns.
Even today we feel compelled to reunite separated pairs, as is materially evidenced in the plethora of single gloves adorning fences and trees worldwide: lost by their owners and displayed by their finders, in the hope that they will one day be reunited with their counterparts (Bissell 2009, 109-10). Altruism may have motivated such actions, but single gloves appear in such contexts far more prolifically than other easily lost items, such as hats, scarves and umbrellas. Indeed, this phenomenon has engendered a Flickr group entitled 'Lost Gloves on Fences' 4 and has culminated in a number of assemblages, whereby particular fences come to be known in the local area as the place for depositing/finding a lost glove. Examples of such single-glove assemblages include Cotham Hill, Bristol, and Laugavegur, Reykjavík, Iceland ( Figure 6). This seemingly innate desire to reconcile two halves of a whole (whether shoes or gloves) may well have been exploited by the concealers of shoes, hoping for protection from malevolent forces. Certain measures were indeed taken to confuse and disorient spirits, such as the burial of spiritually polluted corpses at crossroads (Johnston 1991, 217-18) or the labyrinthine medieval funerary processions designed to prevent a ghost from re-tracing its steps and returning home (Gordon 2013, 85;Wilson 2000, 297). Perhaps it was believed that, if a spirit encountered a single shoe concealed in a house, they would be so confused by the absence of its counterpart, and so occupied with searching for it, that they would be dis-tracted from their original quarry: the occupants of the house.

The other shoe sacrificed
If cognitive stickiness was the objective of the depositor, then the absence of the other shoe was central to the act of concealment. Therefore, rather than simply being disposed of or stored elsewhere in the house, the other shoe may have been actively destroyed or irretrievably deposited-which would again account for why no concealed single shoe has been matched with its counterpart. However, active destruction or irretrievable deposition of the other shoe for the purpose of increasing the cognitive stickiness of the concealment would have been a ritual act in itselfand begs the question of how it was destroyed or disposed of.
One theory is that the other shoe may have been consigned to water. Swann (1996, 65) reports the words of a local Hertfordshire woman who claimed that 'when discarding a pair of worn-out shoes, one should go to water, one to fire, for good luck'. This 'go to water' could have involved deposition in a spring or well (see above), while 'to fire' may fit with the number of concealed shoes (roughly 22 per cent according to Swann's figures: 1996, 123) secreted in fireplaces, hearths and chimneybreasts. It could also fit with the significant proportion of objects concealed alongside shoes which are associated with heat and fire: clay pipes, candles and candlesticks (Swann 2016, 128). Would this signify that their counterparts were deposited in water? Perhaps, but it may instead signify that the custom of consigning to fire outlived that of water, and that some shoes, rather than being placed in close proximity to fire, were actually placed in the fire. After all, it was the smell of burning leather that was believed to be particularly repellent to supernatural forces (MacCulloch 1910, 202;Radford & Radford 1948, 306). Perhaps this is what became of the other shoe: it was incinerated.
Whether the other shoe was consigned to fire, water, or some other irretrievable location, it is not a stretch to claim that it was 'sacrificed'. Active destruction or irredeemable deposition signify the sacrifice of an object according to the definitions of, for example, Bradley (1990, 10), Insoll (2011) and Robertson (1974, 18). However, it is the author's opinion that regardless of what became of the other shoe, its initial separation from its concealed counterpart signifies sacrifice. As posited above, the permanent separation of a pair of shoes -whereby the whole is broken by the removal of a part -is a form of fragmentation, Ceri Houlbrook Laugavegur, Reykjavík, Iceland, 2016. (Photograph: author.) which (in rendering the utilitarian value of the shoes moot) is a form of sacrifice.

Figure 6. (Colour online) A single-glove assemblage on
Although the author is reluctant to view the custom of concealment as a simple survival of a past practice, this could fit with the theory of the concealed shoe being a foundation sacrifice (van Driel-Murray 1999, 137). The practice of burying an offering beneath the foundations of a structure to ensure, for example, luck, has a long history in the British Isles (cf. Rushen 1984, 33), with Ó Súilleabháin (1945 theorizing that the earlier custom of sacrificing animate beings (humans, animals) was gradually replaced by the sacrificing of inanimate objects-and the shoe could indeed be one such inanimate object. Merrifield (1969, 102) suggests that the practice of concealing shoes started as a builder's custom; builders secreted shoes within the areas of structures they had built or renovated (for example, immuring them in walls), in the fashion of foundation sacrifices.
Whether or not we are convinced that concealed shoes constitute a form of foundation sacrifice (the author remains wary of this interpretation), it is still possible that some notion of sacrifice was considered integral to their efficacy. This theory is especially strong when considering the secular origins of the concealed shoe, which would have begun its life as footwear and only later been ritually recycled as an apotropaic device (Houlbrook 2013). The transition from a utilitar-ian to a ritual context requires a rite of passage and the creation of a new self-identity, necessitating the destruction of the old one (Brück 2001, 157). For objects that were initially created and used for secular purposes, this destruction is often literal, with depositors deliberately breaking-i.e. killing (Pollard 2008, 55)-their deposits in order for them to be 're-born' as ritual objects. As Lucero (2008, 192) writes, 'Objects made expressly for ceremonial deposition were never animated and thus did not have to be terminated or killed. Goods people used in life, in contrast, had to be killed before deposition because of the forces they personified'.

The other shoe concealed
Another possibility is that both shoes of the pair were concealed, but in different locations. In the early modern period, people were accustomed to making a little go a long way and were unlikely to dispose of something that was still serviceable. During the period in which most of these concealments were being made (the eighteenth-nineteenth centuries), shoes were expensive items, a pair costing on average the equivalent of a week's wages (Swann 1996, 59), and it is not difficult to imagine that the depositor of a concealed shoe would have been reluctant simply to dispose of its counterpart. However, so far no pair of matching The Other Shoe shoes has been discovered in different locations of the same house. It is more likely, therefore, that if the other shoe was concealed, it was concealed in a different building, perhaps that of a relative, friend, or neighbour.
This sharing of a pair of concealed shoes may have been more than a simple matter of thriftiness; it could have had some ritual significance in itself. Returning to the notion of a pair of shoes being separated as a form of contract, Evans et al. (2016, 245) relate it to the Roman custom of tessera hospitalis, whereby an object was halved, fragmented, but rather than one half being retained and the other offered to a divinity, both were kept by two parties as a symbol of their bond. Earlier examples of such contractual fragmentation may exist from the Middle Neolithic; for example, the small clay models of 'split' legs, consisting of only the right or left leg, found in the northern Peloponnese. Lauren Talalay (1987Talalay ( , 1993 disputes the traditional interpretation of these 'split' legs as 'stray pieces' of figurines. She argues instead that each leg was probably originally attached to its matching half, but that these pairs were designed with the intention of being easily broken apart. They were, she believes, made and employed as social and economic contractual devices or identification markers, serving to symbolize agreements, obligations, friendships, or common bonds (Talalay 1987;1993, 45-6).
A modern-day equivalent is that of the 'friendship charm': two pendants shaped as matching halves of a love-heart and worn by two individuals as a declaration of their friendship. In archaeological theory such objects engender a process known as enchainment, whereby relations are formed and mediated by objects between people over space and time. Chapman (2000, 6) describes the process of enchainment as follows: 'The two people who wish to establish some form of social relationship or conclude some kind of transaction agree on a specific artefact appropriate to the interaction in question and break it in two or more parts, each keeping one or more parts as a token of the relationship'.
The fact that no concealed shoe has been matched with another concealed shoe is not evidence against this theory. Many concealed shoes undoubtedly remain concealed, not yet discovered, and it is certainly possible that one shoe of a pair can be uncovered and recorded, while its counterpart remains secreted away. Many other concealed shoes have likely been disposed of. One shoe of the Otley Cache, Yorkshire (Figure 4), for example, was discarded before the other five shoes were found, and the author has spoken with a number of finders of concealed shoes who have admitted to having thrown their finds away before recognizing their significance. Were these disposed shoes the counterparts of other concealed shoes?
Even in the unlikely event of two matching, separately concealed shoes having been discovered, retained and recorded, it is still doubtful that a researcher would recognize them as a pair. Despite the countrywide range of Northampton Museum's Index, there has, as yet, been no production of a database of concealed shoes complete with photographs which could be utilized in the attempt to match singles. More work clearly needs to be done on the compilation and visual presentation of accessible data in order to establish whether or not the other shoe was itself concealed, and the author hopes to contribute to such a collection of data.

Conclusion
Readers of this paper will undoubtedly have observed that it offers more questions than answers, a fact that is unsurprising considering the enigmatic nature of its subject matter. The concealed shoe is an elusive thing itself, its counterpart even more so. There is certainly scope for some of the questions raised here to be answered-but greater resources would be needed. Raising awareness of the significance of concealed shoes, and of the importance of checking for them in their common locations whenever opportunity arises, would increase the number of finds reported. Greater awareness could also ensure that adequate details are recorded and photographs taken. The production of a countrywide-or, indeed, worldwide-database containing comprehensive details of shoe finds, together with photographic images, would allow for firmer conclusions to be drawn regarding the fate of the other shoe.
However, even with these resources available to us, it would still not be possible to state where the other shoe ended up and why-because there is no the other shoe, but many other shoes, thousands in fact, and we must account for the very likely possibility that they were treated differently by different people at different times. The author has written previously on the 'mutability of meaning', observing that while 'participation in folk customs tends to be formulaic and ritualized … the reasons behind participation and the "meanings" ascribed to the custom will be as varied as the practitioners themselves' (Houlbrook 2014, 41). Perhaps some people consigned their other shoes to water, others to fire, whilst some disposed of them as rubbish and others gifted them to relatives, friends, or neighbours, to be concealed elsewhere. This would also account for the exceptions to the single-shoe norm: for example, the six cases of concealed shoes recorded in this paper's sample that comprised pairs. Clearly some concealers, albeit the minority, did not consider the separation of a pair of shoes central to the custom. Were their motivations distinctly different to the concealers of singles, or is this simply more evidence for the mutability of meaning?
Customs also vary depending upon time, with any original 'meanings' being morphed or forgotten over the many years-centuries-they are observed (Mills 2008, 84;van Driel-Murray 1999, 136). The treatment of the other shoe may have changed significantly over the generations. Or it may have stayed the same but with altered, or even loss of, 'meaning', perhaps falling into that category of deposition described by Pollard (2008, 45) as 'routinized and largely unconsidered'. Perhaps the concealers of shoes did not think too deeply about why they observed such a custom, and therefore may not have considered why they treated the other shoe in the manner that they did. This manner may be elusive to us today, but the study of the other shoe still demonstrates that the absent part of a fragmented object can be very much present in its agency. Returning to Meskell's argument (2008, 237): 'just because something is not on public view … does not necessarily mean that the object is not working'. Notes 1. It is unclear whether this absence of written testimony was an incidental or integral aspect of the ritual of concealment (Eastop 2015, 137). Perhaps, as Swann suggests (1996, 67), 'the secrecy continually encountered suggests that the superstition, if disclosed, ceases to be effective'. 2. The corpus of literature surrounding post-medieval apotropaic devices is vast, but for broad overviews, see Davies (2015); Easton (2015); Hoggard (2004); Hutton (2016); Merrifield (1987). 3. As shoes in Britain were constructed as straights until the mid nineteenth century, there is no way of discerning if there was a preference for concealing the left or right shoe. Following this date, there appears to be little preference: of the 58 catalogued concealed shoes from Nant Gwynant, Gwynedd, 22 were left, 13 were right, and 23 unknown. 4. https://www.flickr.com/groups/75145226@N00/ (accessed 8 October 2016).