Peasant in a Bottle: Infrastructures of Containment for an Italian Wine Cooperative

ABSTRACT On the market for quality wine, quality is generally defined by the liquid’s connection to a time, space, and person(s) of origin. This connection is achieved through material-discursive versions of terroir. While the consequences of this aesthetic regime have been well-studied, little attention has been given to the artefact through which the connection is enacted: the bottle. This paper, based on fieldwork with small-scale wine producers in Italy, asks what difference it makes for these artisanal producers to bottle their wine themselves. The analysis hones in on the bottle as a container embedded within an infrastructure of containment, and emphasises how the bottle as infrastructure re-orders the space–time of circulation. The result has implications for container-mediated processes of production-exchange-consumption everywhere.


Introduction
Chests, especially small caskets, over which we have more complete mastery, are objects that may be opened.When a casket is closed, it is returned to the general community of objects; it takes its place in exterior space.But it opens!. (Bachelard 1994: 85) Early evening, late summer.The sun has descended behind the hills that surround the cooperative, and I stand here in its afterglow.Next to me, Anna looks as tired as I feel.She passes the pouch of tobacco.With our backs to the warehouse building where we have been working since morning, the land around us presents a view dotted with stone houses, a smattering of vegetable gardens, and a church tower.Below us, in the receding light, we can see some of the vines that provide the cooperative with its income.'I am really sorry,' Anna says, rolling another cigarette, 'but we really needed somebody to help out.This is not what we want to show people.'What Anna was referring to was nothing to do with the vineyards below, and the grapes still suspended in the air; it was about grapes that had already been harvested and vinified.Since morning, eight of usparticipating for either the full day or part of ithad been loading bottles onto a conveyor belt that carried them into a machine, from which they emerged on the other side filled with wine and with a stopper in place at the neck.The wine itself had been pumped through a grey plastic tube connecting the machinerented and brought by van whenever the time for bottling comes aroundto one of the fermentation tanks of the cooperative, and into the bottles ready for us to stow away in the warehouse.If not for the (not infrequent) malfunctioning of the machine, we would have neither smelled nor seen nor tasted the wine.Fortunately, the occasional error had the felicitous effect of leaving a bottle uncorked, presenting an opportunity for a drink to pass the time.
The bottles would then remain in the warehouse until the time came for attaching their labels and packing them into cardboard boxes (standardised to fit either six or 12 bottles).These boxes were then stacked onto standard-sized wooden pallets, and the pallets were deposited in the hold of a vehicle that would take them back onto the asphalt roads by which the empty bottles had first arrived at the cooperative.But all that was for later.Today, we had been working diligently, two people at each end of the machine.Apart from the occasional matter of filling new corks into a funnel on top of it, it was a day's work as varied as the 750ml standard bottles we were endlessly loading and unloading.
*** Early evening, late spring .Up in the crown of an ancient oak, I stand amongst the rustling leaves and the cooling breeze, exhausted.A worker with whom I have spent the day in the vineyards has brought me here, and together we have climbed up to a platform hidden in the tree.Now, we stand overlooking the vineyard of a neighbouring producer, who (I have been told) owns significantly more land and operates in a significantly less 'artisanal' manner.We see nobody, but from somewhere in the sea of green foliage below, we hear the sound of work still ongoingthe mechanic buzz of hand-held tools used to tie up the sprawling canopy.A few moments of silence, as we try to spot the workers.Then the man standing next to me speaks.'It is very different how these people work.Very different from us.They always have to be fast, always running.Not enough time to work with caution.One thing is that they are a large cantina, where the people are always coming in and out.'A pause.Then: 'But it also makes such a difference to sell your wine in a bottle.Rather than bulk [sfuso].To have a bottle, with your name on it … is something that gives you a sense of pride.It means that you can't work in just any old way.'The sun is descending over the hillside.I think back to my conversation with Anna, many months before.The act of putting wine into a bottle.On the one hand, it 'makes such a difference'; on the other, it is 'not what we want people to see'.A question begins to take form.*** In summer 2017, I had come to Italy to conduct research with small-scale producers of artisanal wine, and the Piedmontese cooperative where the first bottling took place was my primary fieldsite.Growing grapes and making wine is not the only economic activity of its twenty-something members.But despite the work put into agricultural produce such as grain, fruit, and meat, the wine is indispensable in a way none of the other products are.As one worker put it: 'The wine is the linfathe sapof the cooperative!'.'It is the wine that has saved [us].The only thing that goes plus' the man would say.'You can put a lot of effort into making the best grain in the world, and nobody will pay you another cent for it!'The same held true for all the farms I would visit on the peninsula, from Calabria to the Marches and back to Piedmont.
The scenes described above both depict the practice discussed in this paper, and raise the issue it seeks to address.Bottling 'makes such a difference'.Yetand this is significantnone of the winemakers take any interest in it.'It is a bit … mechanical,' one producer explained, when I asked why.Then there is 'that incredible noise!' the machine itself emits, as another farmer put it when responding to the same question.Not all sentiments were negative.'I like it!' one woman said of bottling.'It's like … I go into a sort of trance.'Such a mental state, however, is contiguous with the experience of repetition and isolation which most people spoke of in critical terms; as the same woman clarified: 'Well, I could still only do it for a day or two.'This paper sets out to explain why producers bottle nonetheless.I argue that while bottling is not 'what [producers] want to show people', it is what enables them to display that which they do want people to see: their land and labour.In so doing, they create opportunities for harbouring 'a sense of pride' in both.Furthermore, by articulating this display on a market organised on the principle of terroirwhich I return to belowthey gain the means by which they reproduce themselves as proprietors of land and labour to take pride in to begin with.
The means in question come in the form of money, and the wine they make is a commodity.Yet commoditisation, as Hart (1982: 40) observes, 'has many concrete manifestations in our past and present and may take several courses in the future'.For decades now, anthropologists have explored how the social life of commodities tells more complex stories than allowed for in the economists' conception of 'an article or a service which is produced for exchange, where exchange is between 'universal others' […] who have no relation with each other except in so far as they exchange these commodities' (Stirrat 1989: 94).Informed by the notion that 'commodities must be not only produced materially as things, but also culturally marked as being a certain kind of thing' (Kopytoff 1986: 64), anthropologists have traced how things circulate in and out of variable commodity forms.Such inquiry, however, is not sufficient for grasping the alleged 'difference' that bottling makes for the commodity of wine.By dichotomising material production and cultural marking, such 'biographical' inquiry tends to overlook two things: the material requirements for the 'cultural marking' of different commodities, and the ways in which categorical transformations are accompanied by 'circulation' of the kind that moves entities through physical space and time.It is by attending to these latter dimensions that we learn how bottling 'makes such a difference'.
In this paper, I conceptualise the relation between 'circulation' in the conventionally economic and the spatio-temporal sense through the concept of infrastructures of value.This approach builds on Graeber's (2001) notion of value as the incorporation of activity into a meaning-bestowing relational totality of action, butas Christof Lammer and André Thiemann argue in the introduction to this special issue -pushes anthropological value theory to unpack how the actual link between actor and whole is established.When turned towards networks of consumption-productionexchange, the concept of infrastructures of value foregrounds the enduring structures of material mediation, which compose these networks in ways that tend to fade from view for those who operate within them. 1 The analysis presented here thus follows a theoretical trajectory from a concern with 'the forms or functions of exchange' to 'the things that are exchanged' (Appadurai 1986: 3), extending it further to the tangible matter of these 'things' and to infrastructures understood as 'matter that enable[s] the movement of other matter' (Larkin 2013: 329).It may be that '[v]alue […] is never an inherent property of objects, but is a judgment made about them by subjects' (Appadurai 1986: 3).However, an examination of infrastructures of value can transcend the subject-object dichotomy still implied here, propelling further efforts to analyse how objects and subjects are brought together in larger wholes wherein judgments are made.In the context of this paper, I show how the material properties of containers must be considered in any effort to explain how acts of bottling compose economic situations that are congenial to the interests of the winemakers.
The paper proceeds as follows.The following two sections begin by addressing the structures of 'value' relevant to the actors at the centre of this study in respect to who they are and what they want wine to do for them.The third section turns to the 'infrastructure', asking what it is about bottling that enables wine to perform the task set by these producers.The fourth section returns to ethnographic exposition in order to display what the preceding discussion enables us to recognise as infrastructures of value.

Wine Within Exigencies of Scale
This paper is concerned with Italian farmers who grow grapes and make wine.But what is it that they want the wine to do for them?Why do they produce wine to begin with?In the words of one producer, the answer has to do with what he referred to as the 'vicious circle' of industrial agriculture.'That is how it works, no?' this broad man in his late forties mused as we were making a tour of his vineyards.'It encourages you to produce a lot; then the prices go down … and when the prices go down you need to produce even more.'For him, however, there is now a means of escape from these dynamics.'With this wine,' he continued, 'I go my own way.Independent of this circle.In any case, I think it is probably the only possible route for a small-scale producer like me.'The same sentiment is echoed by winemakers across the peninsula: 'Wine provides a return'; 'Wine makes it possible to survive'; 'It pays off'.Or, in the more elaborated words of a neo-rural emigrant from the city, now in her forties, more elaborated: when I asked why she had abandoned her original vision of selling a mix of agricultural products: 'Well, partly because of an interest in wine.But also because the wine provides a return.We have only five hectares, so wine is the only thing we can possibly live from.With five hectares of grain … we would have nothing!' The primary interest of these producers is to subsist as small-scale farmers, and wine is what enables them to do so.We encounter this concern with small-scale farming in a country shaped by a century of oscillating scales of agriculture: first from large to small, as a post-war reform partitioned large estates (especially in the South) in the hope of creating a stable landowning peasantry; then from small to large, as the partitioned holdings proved too small for their proprietors to survive the incrementally diminishing returns on each unit of production, which mechanisation and price-based competition brought about (see Silverman 1971;Davis 1973;van der Ploeg 2012).The result: A vast exodus of the rural population to the cities, and the emergence of 'peasant-workers', who subsist in agriculture through complementary income from industrial employment (Holmes 1989).Large-scale farming gained ground on small-scale, and those who remained small could not do so as the full-time farmers the wine-producers aspire to be.
Wine now provides a 'possible route' for the small-of-scale.Crucially, this is not due to any qualities inherent in the wine itself.The trajectory of Italian wine throughout the twentieth century makes this abundantly clear.In the period leading up to the 1980s, the productive volume for Italian wine doubleda trend that culminated in 1982, when Italy stood for 39% of the global volume of wine, but only 20% of the global value (Corsi et al. 2018: 160).This was decisively not to the advantage of small-scale producers.Even as volumes increased, the number of grape-growing farms diminished, and the wine market was characterised by 'large branded firms and downward price pressure on growers' (Carter 2017: 494).In this context, wine offered the small-scale producers no escape from any vicious circle.
As for how wine gained its present powers, the producers themselves provide an apposite account.'There are several reasons that things went well,' said an older winemaker.'The first is this: the first wine we ever put in a bottle, we released in '85.Just before the methanol scandal hit.This made it so that people turned to us, thinking that perhaps this could be something [worthwhile] … ' In 1986, the scandal in question took place.Certain Italian wine merchants adulterated their wine with methanol, leaving 23 dead from drinking it.Even if this was not the sole cause of changes to comethere is also a connection to a broader global shift towards artisanal products the significance of the scandal is well documented (e.g.Barbera and Audifredi 2012).With the events of 1986, the quantity-based market for Italian wines collapsed.And when it recovered, 'quality' would play a central role, while quantity decreased.In 2000, yields were 23 per cent lower than in 1985.But despite falling yields, the wine sector now shows the best economic performance of any agricultural sector in Italy, producing a total value second in Europe only to France (ISTAT 2018).Sixty per cent of the Italian turnover from wine is still in the hands of 136 companies (Corsi et al. 2018: 167).Nonetheless, there is also a stable niche for 'a galaxy of small quality-oriented integrated wineries' (Corsi et al. 2018: 167), such as those discussed in this paper.
In sum, wine makes a difference only due to a specific recontextualisation (Myers 2005;Cavanaugh 2007) which it has undergone under specific historical circumstances.Wine empowers small-scale producers to 'go [their] own way' insofar as it draws an economic return upon considerations of quality (see Karpik 2010).Winemakers today compete through assertions of quality, rather than by pressing costs of production, and this serves their efforts to escape what they regard as the 'vicious circle' of industrial agriculture.
But what is the connection of any of this to bottling?First, when it comes to the historical shift from quantity to quality (and large-scale to small-scale), there is a scarcely noted historical correlation (see Corsi et al. 2018Corsi et al. ): in 1947, bulk (unbottled, tank) , bulk (unbottled, tank) wine represented 87 per cent of the export volume of Italian wine, and 73 per cent of the value.In 1970, when Italian wine exports took off in earnest, the proportions were 82 per cent volume for bulk wine, for 61 per cent of the value.In 2014, by contrast, 27 per cent of export volume was in bulk, for only 8 per cent of the value.Bottled wine, during the same period, moved from 11 to 61 per cent of export volume, and from 24 to 75 per cent of export value.Not all of the bottled wine exported would count as 'quality wine' by any measure.However, with the 'quality turn' that took place in Italy, there was also a turn towards the bottle.Considering this connection as pertaining to infrastructures of value pinpoints the dynamics through which bottling 'makes such a difference' for small-scale winemakers today.

Value as Place
'Quality' can be identified by way of any number of properties.Not all such identifications would mean much for practices of bottling.One that does, however, is at work amongst the wine producers discussed in this article.One man captured both the aesthetic regime of the market and that regime's operations with precision: 'It is territoriality [territorialitá] that has saved the cooperative.This is something we learned from the French … and it is wine that has saved territoriality!'The statement, of course, employs the Italian version of a word the Italian wine producers otherwise speak almost exclusively in the original French: terroir.
The term terroir, often glossed as 'the taste of place' (Trubek 2008), invokes an idea of 'wine with a taste unchanged since time immemorial' (Demossier 2011: 690).A considerable body of academic literature, however, traces the modern history of the concept itself (Guy 2007; Fourcade 2012; Demossier 2018), the extensive legal-scientific infrastructures used to evidence its unique 'natural' features (Ana, this issue), and how terroir has diffused from its original exclusive association with specific French wine to other places (Jung 2014) and products (Paxson 2010). 2 While similar constructs are now associated with a range of Italian agricultural products, from salami (Cavanaugh 2007) to pork fat (Leitch 2003), cheese (Grasseni 2016) and beer (Fastigi et al. 2015), wine is the historical home of the model.
The above claim about territorialitá, therefore, recognises that the economic significance of wine hinges on construing value from the relation between the final consumer and the place and time from which the primary materialthe grapesoriginates. 3In this context, terroir provides small-scale producers a way out of the alleged 'vicious circle' due to its valorisation of particular locations, as well as its assertion that quality is nonscalable in the sense that winemaking cannot be scaled up without transforming the propertiesspecificity of person, time and placethat define what is good about the resulting product (see Tsing 2015).
Significantly, the connection between bottles and terroir predates the quality turn in Italy.As wine scholar Steve Charters (2006: 51) asserts, 'there are two types of wine in the world'bulk wine and premium wine, associated with large tanks and smaller bottles, respectively.This distinction also maps onto a division between wine that is indexed by specificity of place, and wine that is not (Murray & Overton 2011).Already in the 1660s, when the earliest wine sold under the name of a single estate appeared, it did so in direct conjunction with the appearance of new mass-manufactured wine bottles some three decades earlier (Howland 2013).While far from all wine found in such bottles aspires to be recognised as 'good' in terms of its association with a single vineyard, it is almost exclusively in such bottles that you will find wine that does.
To understand why this is so, we need to turn to the infrastructural requirements presupposed wherever terroir is enacted.As Demossier puts it, terroir is a matter of 'experiencing being-in-location through a displaced and mediated consumption of singularities' (Demossier 2018: 87).'Singularity' is a matter of an identifiable place, and I have already discussed how scholarship traces the development of such valorisation.But what about the operations that actually allow this singularity to be mediated and displacedas this singularity?Even as terroir is construed by a range of evidencing infrastructures (Ana, this issue), the connection by which a consumer can experience being-in-location presupposes that the grapes of the wine are literally from the location in question (Nowak 2019).Thus, the wine needs to circulate along trajectories where its connection to the specific somewhere of its origin is upheld in destinations located generically anywhere. 4By attending to the infrastructure of value that makes this possible, the connection between terroir and bottles becomes clear, and so does the reason that bottling 'makes such a difference'.

Bottles as Infrastructures of Containment
What first interested me about bottles was the fact that no winemaker takes any interest in them.Bottling is not something they speak of in the same way they might talk of their other labours.Nor have I ever heard anybody speak admiringly of somebody's 'well-bottled' wine, the way they speak of a 'well-made' wine.But as bottles are artefacts omnipresent in the background of their working lives, a direct inquiry is warranted.In the preceding sections, I have already provided context: the most crucial 'difference' wine makes for these (small-scale) producers is an economic difference, this difference is made by wine recontextualised to a domain for quality competition, and this specific domain is governed by the principle of terroir.In the present section, I argue that there is an intrinsic connection between this quality regime and the integration of bottling infrastructure into the wine-producing farm.The starting point for the argument lies with the key artefact employed in acts of bottling: the bottle itself.
Bottles come in a number of shapes, sizes, and materials.Wine bottles, however, are readily recognisable in retail environments all over the world.The material is conventionally glass; the standard size is 750 ml; the shape is rounded in a way that fits the grip of a hand.Not all wine bottles are alike, of course.Aside from variations in volume, the shape of the bottle may communicate something about the content.While there are plenty of exceptions, the vast majority of wine bottles fit a threefold typology.As one Italian wine producer remarked, 'There are only three kinds of bottles: the Bordeaux bottle, the Burgundy, and the Alsace.Any other style is just flashy nonsense!'Since the eighteenth century, these three wine regions have been associated with distinct bottle shapes: the Bordeaux straight and firm, the Burgundy rounded and broad, and the Alsace tall and narrow.While the rejection of the 'nonsense' shows that other styles do in fact exist, the sentiment thus expressed is crucial for the argument on infrastructure developed here.Whereas some wine producers seek to distinguish themselves by means of bottle shape, the wine producers discussed in this article all dismiss such strategies.They may speak of the value of their special soil or arduous labour, but not their bottles.If bottling is said to 'make such a difference', then the bottle's role in making the difference remains at an infra-structural level, where it often falls out of view, even as it is indispensable for having made present whatever this view overtly regards (Peters 2015a).Or tastes, in the case of wine.
The acknowledged 'difference' that this paper inquires into thus does not hinge on how one bottle differs from another.Instead, it is about how bottles perform the function for which the producers requisition them: containment.This puts our inquiry in an ambit seldom explored in the literature.Such containment is enacted by artefacts that possess an inside, an outside, andconventionallysome feature for inserting and extracting content (on the self-containment of raspberries through deep-freezing, see Thiemann, this issue).In academic accounts of material culture, these artefacts have a propensity to disappear in favour of histories focused on more actively intervening artefacts (Mumford 1934: 11-12;Leroi-Gourhan 1993: 133-134;Peters 2015b: 140). 5Recently, however, progress has been made in this respect.Shryock and Smail (2018: 1) have delineated a 'container' as an artefact able to 'hold something else inside itself for an indefinite period of time, isolating the contents from the give and take of the world outside' (Shryock & Smail 2018: 1), and Robb (2018) suggests that four things follow: such content is preserved as what it is, it transcends time, transcends space, and it can be ordered in a specific ways. 6The value of wine itself is construed within a broader infrastructure composed of labels, pallets, vehicles, roads, categorisations, information and stories (on the link between containers and information infrastructures, see Lammer, this issue).If we are to understand the difference bottling makes, we need to consider how these capacities of containers are put to work, with delineable effects, at different points within this infrastructure.
Let us return to the introductory vignette in order to consider containment along the 'social career' (Kopytoff 1986: 66) the bottlers expect of their bottles.First, the content.Whatever its accidental properties, for the argument made here it is indispensable to note that wine is invariably a liquid.Also invariably, human hands are therefore not wont to grab hold of it in order to transport it any distance in time or space. 7This is the reason for the tanks that hold and preserve the wine over the period of its fermentation, as well as the tubes that transport it the short distance to the bottling machine.Once completed, the act of bottling puts into effect a key transition.Isolated from 'the give and take of the world', the wine is preserved for an indefinite period of time.The bottles are then ordered by means of their deposition into the cardboard boxes, themselves ordered through stacking onto a standard-size pallet (see Orlandi 2017).At this point, the wine is loaded into a container taking the form of the hold of a lorry, whence it is transported into spaces far transcendent of the farm where the grapes had been grown.The expectation for the further social career of the wine-commodity (by which these grape-growing farmers make their living) is that each layer of containment will be peeled awayfrom lorry to pallet to cardboard box to bottleand into the glass of the final consumer.At this destination, the liquid will again enter into exchange with the give and take of the worldconcomitantly losing the container-induced ordering that might have preserved and transported it yet further in time and space.
Recall the requirement imposed by terroir: that it allows wine to index a location only as singular as that from whence its material substance originates.Along the chain of transport just described, that specific locational projection is achieved for the winemaker by the bottles.Note the sequence of sub-packaging.Containers also have the capacity to contain other containers.With the industrial revolution, this capacity was increasingly put to use (Bevan 2014), associated with an increasingly global system of transport (Klose 2015).Bottling, here, is a step which turns the liquid substance into something displaceable as a solid, thus aligning it with the operations of a transport system capable of bringing it (almost) anywhere in time and space.And, as Klose puts it when referring to this system, the 'principle of container transport [lies] in conceptualising the container as a metacontainer, whose cargo is not of concern, as long as you do not find yourself at the beginning or the end of the transport chain' (Klose 2015: 315-316).As the bottle separates its content from the 'give and take of the world', the properties of the contenteven its liquid stateno longer make a difference.
Precisely this separation from the surrounding world responds to the second requirement of terroir.As terroir, wine is valued due to its purported capacity to tell the story of its origin.But 'origin', here, is not the miles of concrete the wine traverses to reach its destination, nor is it the people and activities with which it is involved along the way.Origin is place of cultivation.And here, the principle of container transport works in reverse, by creating the possibility of a relation of 'concern' between those who do find themselves at the beginning and end of the chain.As the content is 'not of concern' for the commodity's handling along the course of transport, its handling makes no difference in respect to its terroir.
It is by attending to the heterogeneous elements involved in infrastructures of value from liquids to pallets to ideasthat we can appreciate the significance bottling has for terroir.But such an understanding is incomplete without appreciation for how bottles operate as containers.As a container, the very structure of the bottle makes it an infrastructural link, a mechanism that connects the point where it is sealed to the point where it is opened.Whatever its shape or colour, the standard-size bottle discussed here is meant to be opened by a final consumer.And by integrating the act of sealing into the grape-growing farm itself, the winemakers ensure that they themselves are made present at that final point, irrespective of whether it is Turin or Tokyo.
The next section again turns to the historical and everyday contexts of Italian wine production to show these infrastructures at work.

Containing Differences
If bottling the wine is a choice farmers make, this is only because there are other ways of containing it.While an infinite variety of containers are capable of containing liquids (say, the amphora), the containers available for wine in present-day Italy fit a threefold typology.A brief excursion to a debate from the immediate historical past can help delineate the categories and their differences.Here, we turn to two first-hand witnesses to the twentieth-century recontextualisation of Italian wine: Mario Soldati  and Luigi Veronelli .As journalists and food and wine critics who have exercised an unparalleled influence on gastronomic culture in Italy, both were engaged in the same endeavour: to safeguard the remnants of the vanishing 'peasant world' of Italy, and to do so by means of documenting and preserving peasant food and drinkin particular, wine.They were both aware that the burgeoning quantity regime of wine (outlined above) provided the small-scale farming peasants very little in return for the grapes they sent away in large tanks.Moreover, illustrating that the propensity of infrastructure to fall into the background is always a reversible process (Star 1999: 380), both looked to other containers for answers as to what might make a difference for the peasants.Yet precisely here, they found themselves in disagreement.Where Veronelli believed the peasants ought to put their wine in bottles, Soldati championed the demi-john [damigiana] (Lorigliola 2017: 39).8  Both men construed the tanks of the industry in opposition to the bottles or demijohns of the peasantry.The ways in which the latter two might serve the small-scale farmers (Veronelli and Soldati believed) lay in their connection to distinct modes of circulation.Soldati championed an older model.The peasantry of Italy has long made wine of its own; but seldom was this wine destined for the market (Corsi et al. 2018), and if so, then primarily to nearby towns (Federico & Martinelli 2018).Authentic peasant wine, for Soldati, is (demi-john) wine that circulates through social networks of friends and family, never treated as a commodity (Lorigliola 2017).Veronelli, meanwhile, hoping that (bottled) wine would allow the peasants to gain the revenue by which to keep hold of their own means of production, favoured a new form of 'territorial' recontextualisation.In this recontextualised circulation, wine would remain connected to the original producer even while circulating as a commodity on a global market, in the manner associated with terroir.
The works of Soldati and Veronelli still line the shelves of many winemakers, and quotations by the latter are in common parlance.The significance of the debate, however, has to do with more than the presence of these figures in memory.What it provides here is a 'native' precedent for recognising three forms of circulation, coupled to three distinct containers.These forms, and the recognition thereof, have remained with winemakers up until the present day. 9While they all rely on Veronelli's strategy for their survival, the threefold difference between containers and concomitant modes of circulation is gleaned by attending to how the other container-circuits are not entirely absent.Said one producer, recounting the role of the demi-john in his own viticultural biography: 'We would fill the demi-johns, and bring them to the social centres … where it is possible to be without being regulated.We did not have to pay any taxes, which allowed us to get by.' 10 Corsi et al. (2019) claim that 35% of the Italian domestic consumption of wine is bought in this kind of bulk.But, as shown by the quote above, this wine often flies under the radar, rarely enters any statistics, and seldom travels much further in time and space than the producers themselves.While many producers I know do sell some wine in this manner, it is the mainstay of none of them.Continued the producer quoted above: 'We did this until we started to become more well-known.Then it became too dangerous.Now we sell it in a bottle.The wine is better now … but it is still basically the same wine.But in bottles you can put a higher price on it.'My arboreal interlocutor from the introductory vignette spoke of 'pride' in the wine.While he did not specify the audience in relation to which such pride may be realised, I often witnessed how these producers would compose audiences for each other.'Damn … this is good.How does he do it?Incredible!' said one winemaker who had come for dinner at the home of another, as we opened (and finished) a bottle from a producer at a different end of the peninsula.The host nodded: 'I know.We should call him and tell him!'I do not know if they ever made that call.But the fact that they could have done shows how the bottled wine puts on display a person with a phone number by which to do so.Here, a specific infrastructure of bottles, roads and labels construed a direct connection between 'singular' places and times.
Finally, even the tank system has an occasional presence on these small-scale farms.For example, one day during the harvest, my hosts were heading for a vineyard far distant from the one where we usually worked.Whereas instructions even for the relatively low-skilled work of harvesting grapes is usually meticulous, on this day they were anything but.'Pick everything!Beautiful and ugly [grapes], it all goes to the cantina sociale where they dump a load of sulphites on it anyway!'The cantina sociale, in this case, is the facility to which grape growers in the area sell their grapes for vinification and bottling.Traditionally, they would be paid by volume of grapes alone, and the bottles sold would be labelled with the name of the area rather than their own farms (on the link between property systems and evidencing infrastructures, see Ana, this issue).Hence, the lamentation of a worker coming upon a particularly beautiful row of vine: 'No!All these beautiful grapes for the cantina sociale!What a waste!'The vineyards were not those of the workers, and the grapes were not destined for their own bottles.When working against the background of this infrastructure, the same people who would otherwise 'work with care' in the vineyardsrecalling again the introductory vignettewere suddenly moving with as much haste as our non-bottling neighbours observed from atop the tree.
It is by attending to how containers operate within infrastructures of value that we discover why each mode of circulation connects to a distinct kind of container.Tanks, bottles, demi-johnseach separates inside from outside, and sets content apart from the 'give and take of the world'.But in each case, the sealing and opening of the container happens in a different manner.Even when the wine reaches the final consumer in a bottle, the existence of a stage where grapes are mixed with the grapes of other producers, in a tank downstream from the farm, means that it arrives stripped of the connection to origins as 'singular' as the winemakers themselves.The wine of the demi-john does mediate a strong connection between producer and consumer, but does so because it is typically the producer who seals and opens it for the final consumer who has come for the wine.Thus, its spatio-temporal circulation is narrow.The bottle sealed by the farmers themselves, finally, is the container that can both connect a wine to themselves and bring it to consumers located anywhere in time and space.It does so due to the significance containers accord the act of sealing.It is by thus responding to terroir's valorisation of non-scalable singularity, ultimately, that 'bottling makes such a difference'.

Conclusion
A statement made high up in the crown of an ancient oak raised the question of the 'difference' made when workers of a vineyard identify with the land that they work, and themselves put its wine in a bottle.To explain why these producers bottle, the paper begins by establishing that their wine makes a difference as a product that competes on quality rather than price.It then shows what defines 'quality' in this case, and points out that the quality regime in question requires that wine can move through time and space in a specific manner.The paper thus answers the question by showing how bottles respond to this requirement.This answer focuses on the bottle as a species of container, which shifts the discussion to the ability of infrastructures to reconfigure time and space.Bottles here are a means for showing by hidingfor connecting points of insertion and extraction while occluding the pathway between.This is their role within the specific infrastructure of value that allows the farmers to reproduce themselves as farmers.
The analysis of this infrastructure traces how disparate elementsland, consumers, producers, roads, moneyare materially linked into a larger whole.While it demonstrates the significance of bottles for this linking, it does not locate a singular 'cause', in the sense where bottles could be said to cause the regime of terroir.To account for how the possibilities 'enabled' by the bottles are brought to a particular realisation, their containerhood must be considered in light of how specific containers interlock with specific bottling machinery, legal regulations about packaging, stories, and ideas about how a liquid can be interpreted in terms of space and time (and with evidencing infrastructures which attest to such identifiers; see Ana, this issue).It is likewise this context that determines that the specific container employed by the winemakers is a bottle.The function the bottle performs as container alone does not account for how it differs from (say) a Bag-in-Box.Yet all my interlocutors have been reluctant to introduce the Bag-in-Box, and the one producer who employs them uses them only for their lowest-grade wine.An infrastructural perspective on containers highlights that the different modules of this infrastructure need to be brought in alignment (Lammer and Thiemann, this issue).Potentially, creating links to stories about the demi-johns of peasant communities, and breaking links to stories about the tanks of large-scale industrial agriculture, would enable the introduction of the Bag-in-Box as a new infrastructure of containment for small producers.Recursively, however, we must understand the operations and transformations of such stories in relation to the specific containers through which systems of meaning are enacted.The Bag-in-Box might become the equal of the bottle only because it is a container that, just like the bottle, is destined for a final consumer and might be sealed at the location of the farm of its origin.It is through its ability to pinpoint this latter dynamic that an analysis of containers becomes indispensable for understanding infrastructures of value.
There is an extensive body of scholarship on how an agricultural product 'will be produced, circulate, and be consumed' (Cavanaugh 2007: 151).But where that literature turns to material properties, it tends to do so for the first and final contexts production or consumption.Deploying the concept of infrastructures of value, this paper has turned the analysis also towards the material properties involved in circulation.Containers, solids, liquidssuch materials are encountered everywhere both in agri-food studies and beyond.By attending to the interaction of materials with practices of containment, and by analysing the significance of containers in respect to their part in infrastructural networks, these matters are kept from fading into the taken-for-granted background where they so often play parts hidden in plain sight.Notes 1. Endurance and visibility are essential to the concept of infrastructure used here, yet these properties are relationally defined (and always reversible) in any specific entity to which the concept is applied.2. In its original institutional form (founded with the appellation d'origine contrôlée (AOC) system of 1935), terroir served to help certain French wine regions draw monopoly rents in a context of intensifying global competition (Demossier 2018).This kind of terroir would be of little use to Italian small-scale peasants.Notably, Italy did introduce a system of its own based on the French model: the denominazione di origine controllata (DOC) legislation of 1963.Unlike the AOC, however, the DOC long failed to introduce any 'quality turn' (Carter 2017).3. Furthermore, long before its arrival in Italy, the terroir model already valorised actors of the kind the wine producers aspire to be: peasants, in direct relation with the historical peasantry (Demossier 2011: 696).See Krause (2005) for the significance of the 'peasant' figure in contemporary Italy, and Cavanaugh (2007) and Helstosky (2004) for the more recent indexing of foods to this 'peasant' embodiment of a collective past.See Black (2013) for how the development of 'natural' winea celebration of peasant winewas delayed in Italy compared to many other countries, due to connotations of this recent peasant past with hardship and hunger.4. It is in respect to the destination 'anywhere' that these efforts to enact terroir generate a different spatial dynamic than the efforts to construct localised market circuits described in Grasseni (2014).5.In addition to the fact that the containers' 'productive roles are not necessarily evident in the final product', Zoë Sofia adds a gender dimension to this occlusion (Sofia 2000: 182).An exception to the lack of concern with bottles is found in Miller's (2002) account of Coca-Cola bottles and bottling plants in Trinidad.As the 'value' of the Coca-Cola does not rely on territorial marking, however, its infrastructure of value operates in a manner different from the one explored here.
6. Some containersfor wine, most notably oak barrels (del Alamo-Sanza & Nevares 2018)perform active functions that are evident in the final product.These functions, however, can often be separated from the container itself (as in the use of oak chips to reproduce the olfactory effect of the oak barrel).They are not part of what defines the containerhood of the container.7. Thus, containers are an indispensable requirement for the trade of oil and wine that has taken place across the Mediterranean for some 5,000 years (Bevan 2014).8.The demi-john is a glass container of varying size (usually encountered at volumes of some 50 litres) which retains connotations with the Italian peasantry.Conventionally, the demi-john is a source from which wine is dispensed into smaller containers for everyday use, either by having the demi-john itself at home or by arriving at the point of sale with empty bottles to fill and bring home.9.This threefold typology of circulation is discussed in Cavanaugh (2007).What the debate between Soldati and Veronelli adds is the issue of the container, which does not enter in the same way for the non-liquid product Cavanaugh is concerned with (salami).Italian arrangements focused on 'co-production' (Grasseni 2013) are akin to the face-to-face circulation via demi-johns, but do not enter as an economic mainstay at the farms included in this study.10. 'Social centres' are long-term occupied urban spaces, with communist and anarchist affiliations.