The Calculative Enactment of Certainty Achieving Environmental Data in a Fortune 50 Company Ingmar Lippert∗ 15th June 2013 Facts on environmental goods and bads abound. Science, policy and management produce and enrol environmental facts to mutually construct frames and systems that translate environmental reality into the more or less hybrid forums which do, supposedly, sustainable development. Environmental facts matter. Nigel Thrift (2004, 587) talks of ‘ontological wars, defining [. . . ] how the situation turns up in the first place’. Facts do not merely represent worlds but take part in enacting it. They decide what is certain about our world. Studying the work of fact-making, thus, is key. STS must show how certainty is achieved, not just given, Helen Verran (2001, 45–46) says. Closely attending to environmental facts can result in much wonder. In a recent paper, Verran (2012a) is puzzled about a much cited and ‘spectacular’ number in environmental discourse, Costanza et al.’s (1997) ‘value of the world’s ecosystem services and natural capital’. The number the latter offer, Verran (2012a) finds, does not fit the prominent stage – the journal Nature – that supposedly presents only the most rigorous rule-bound scientific results. To engage with the presence of environmental ‘facts’ I share a vignette of environmental fact-making in, supposedly, rigorous rule-bound corporate environmental accounting practice. The number I trace here is the total distance travelled wot short-haul flights by employees of a Fortune 50 multinational’s subsidiary. That distance was, eventually, translated into carbon emissions and channeled into global discourses and economies of ‘the environment’. Faced with such environmental fact-making both universalists and relativists, despite their different takes, would likely ask: is any of these facts certain? Are they or not? Also auditors asked the company this question and figured: no, even their data was not good enough. However, the facts were emitted into global carbon realities anyway. Why – or how? This paper translates these concerns into an investigation of how certainties are enacted in the doing of environmental facts and data. Studying fact-making is what much STS is about. Opening up in ethnographic detail how quantitative data work actually proceeds is more rare. Lave’s (1988) contribution is prime – offering recognition that every-day calculation may differ from textbook realities and that this is not problematic in itself. Later, drawing on actor-network theory, Callon (1998) translates the recognition that fact and value-making are often practically not distinct towards economic sociologies. What the matter of goods and services is and how it is quantified became more and more an open question. Decisively, what the matter was lost its stable status and became theorised as process, practice and deeply political (Law 2004). Neither quantities nor qualities, least of all their distinction, can now be taken for granted. ∗ National University of Singapore, Tembusu College 1 This paper relates to and brings into resonance three strands of these debates. Qualculation is the first strand. It revolves around the rethinking what happens to rational decisionmaking when qualitative judgement and quantitative computation lose their boundary. This strand primarily draws on the work by Michel Callon and John Law (2005). Using Peircian semiotics is the second move. With his apparatus we can understand signs not as given and stable entities but as relational material-semiotic-cognitive acting (Peirce 1940).1 This strand relates to recent work by Fabian Muniesa (2007) as well as Bill Maurer (2008) and Helen Verran (2010a, 2012b). It follows, third, to open up how the mattering of numbers is effected. This involves engaging with the relations in which numbers are made present to users and how that constrains or flexibilises number use. Reading Verran’s (2001) Science and an African Logic was key for this strand. Together, I understand these strands as generative to investigate the doing of numbers and its ontic effects – which should be a significant concern for sociologies of calculation. With Verran (1999, 150) we approach certainty of numbers. Drawing on her work amongst Yoruba children learning calculating, she proposes: Certainty of numbers is an outcome of the routines by which they are constituted in collective acting, not in their unique capacity to truly represent a foundation in a system of symbols. She develops these considerations in Science and an African Logic and proposes that in the very practices of counting an object not only the number is performed but the object as well. The pure matter of, say, students – when we count them, in her approach is not antecedent to the action but is brought into reality in that very performance. The outcome ‘73 students’ is not a given pure fact but an outcome that is inextricably interwoven with other entities, such as the list of participants, or, indeed, ourselves. Being certain of the count emerges within the doing of the counting, that is in the acting in a relationally structured situation. My investigation is not about counting students, but about calculating a distance. The resulting fact was, and continues to be, translated into carbon emissions. Carbon, here, is not necessarily just a molecule but encompasses the practices that constitute carbon, e. g. via accounting (Lippert 2012b). Doing carbon via flight distances is, correspondingly, just one of many ways of ‘Enacting Environments’ (Lippert 2013). Next, I situate and introduce the ethnographic material I employ for this paper. Subsequently I analyse the particular case of doing environmental data first in terms of qualculation and, then, in terms of Peirce’s (1940)’s triadic semiotics. This allows us to spell out how doing environmental data enacts particular flight and, thus, carbon matter. I then focus on the achievement of certainties that I find to be interwoven with the doing of numbers. 1. This paper positions itself after the phase of in-principle anti-cognitivism. I do not disallow my analysis to recognise the human body as relating to and transforming data/information/thoughts. This does not mean that I treat numbers as mental entities (cf. Verran 2001); but it means that materiality matters and may include the mind. We can expect calculations to be performed in more or less extended and distributed heterogenous relations, involving e. g. brains, hands (and other body parts), pen&paper, spreadsheets, etc. (Lippert 2011; Sørensen 2012). 2 From carbon markets to bookkeeping As a real-world ‘context’ of distances and emissions, think of climate change. Local and global climate change has been on scientific and policy agendas for more than one- to twohundred years (e. g. Grove 1996). Since the 1990s, scientific, economic and policy debate has agreed upon a particular set of analytical methods and concepts to imagine and fight climate change. Key here is imagining climate change as market failure (Stern 2006). That is to say: in hegemonic discourse, actors agree that climate change and greenhouse gas levels in the atmosphere are a function of economic action. With Callon (1998) we can interpret this take as an attempt to partially include the atmosphere in an economic frame. We, therefore, can expect overflows – things that do not fit this frame. To solve this economic question, as it were, climate and industrial emissions merely have to be translated into the market (MacKenzie 2009).2 Callon (2009) considers such translations an experiment. Now, how is something translated into the market? With price tags and with accounting sheets. Determining costs of production requires book-keepers. To cut is short, you can trace climate change easily back to a set of work practices that communicate climate change within companies: environmental book-keeping practices (Lippert 2013). Such practices can be found in nation states’ environmental departments, in non-governmental organisations like the World Wide Fund For Nature or around corporate sustainability management teams. I take you to the latter. To be more precise, my story takes you the environmental book-keeper of a subsidiary of a multinational company, in one of the fifty biggest corporations of the world.3 This subsidiary is located in Western Asia. We enter the office of this book-keeper; I call him Nick (Lippert 2012a). He is an engineer by profession and got a task by his head quarter colleagues: Nick’s task was to report how many kilometres the subsidiary’s employees have travelled on short-haul flights in the prior year, 2008. This is just one indicator. Others exist as well. All these indicators are, in peculiar ways, translated by head quarter staff into carbon emissions. That is, translated into the emissions that the company will report and emit into global discourses of climate change. So, Nick was supposed to report his superiors a fact of distance. The corresponding category of short-haul flights would, later on, be translated into carbon emissions with a different carbon conversion factor compared to long-distance flights. According to standard conversion factors, short-haul flights cause higher emissions than long-distance flights (Lippert 2013, 101).4 The problems start long before we get to conversion factors. Let me set out from here: Nick could not identify anywhere within the subsidiary a fact that he could simply copy into the corporation’s database. The subsidiary simply did not know about the distances travelled. However, Nick was able to know something quite different: he was able to get a fact about the costs incurred for domestic flights, 168,078 EUR. Using the number of the 2. And, criticism of the imaginary that such translation is possible or good abounds (e. g. Lohmann 2009; Gilbertson and Reyes 2009). 3. A note on method: I conducted field work across 20 months in the multinational 2008-10; the workers knew I research them; and I was paid in parallel by the company to support them in optimising their environmental accounting database. Detailed methodological outlines of the study are available (Lippert 2013, 41–57; fortcoming). Names and numbers are deliberately rendered anonymous to protect informants. 4. The conversion factors differ because in short-haul flights the emissions resulting from take-off and airport infrastructure relative to the emissions by a plane flying in ‘parallel’ to earth’s surface is larger than in long-haul flights (326 g versus 180 g per kilometer). 3 costs he performed a particular calculation. This is quite simple maths – but it teaches us much about how numbers are brought into reality as certain – and we can learn about the quality of this certainty. This is the formula: total f light costsdom. f lights ∗ average f light dist.dom. f lights average f light costsdom. f light = total dist.dom. f lights [where dom. refers to domestic and dist. refers to distance] And this is the calculation: 168,078 EU R ∗ 500 km 230 EU R = 365,387 km I had multiple roles during this research: I was doing participant observation in this moment – and I was to support him as a worker – and I wanted to learn about the ways data gets constructed. One contingent effect of these multiple roles was that I proposed to Nick to store with this new fact – the 365,387 kilometres – a comment in the database that would describe this very calculation so users of the database could transparently see the quality of this fact. He hesitated – but acted accordingly. While we could simply rely on the calculation and take it at face value, in the following I attempt to detail how generative it can be to actually open up the work done in the process of calculating. Achieving a ‘calculation’ How has Nick actually achieved to produce this entity that is normally called calculation? What do we grasp by analysing this as a mathematical operation? Conceptualising this set of relations, this movement of signs as a mathematical office operation suggests that all the entities involved in it are unproblematic; we grasp it as a rule-following method, an implementation of the rules of multiplication and division. This understanding misses the practical point of the operation: it was not about solving a mathematical problem but about bringing into reality an entity that before had not existed. Thus, Nick’s practice had a transformative character: it altered the form of how these entities existed; he assembled them in a shared plane in which he conducted the operation. This transformative movement deserves spelling out. First, he mobilised the total cost fact to draw on for the calculation. Where did this fact come from? Picking up the phone, Nick, had called a colleague and received the cost fact on domestic flights for the subsidiary. This is not trivial. While for this particular case he managed to ‘immediately’ access such a cost fact for the totality of the subsidiary, for many other environmental indicators he could only mobilise partial statements. The latter usually took the form of costs incurred for particular consumption types, such as of electricity, gas and water at a particular office site. Normally, relevant office sites were represented partially. Nick then extrapolated the available site-specific consumption facts to the scale of the subsidiary (see Lippert 2013, 88–94). For our story we can safely keep this form of extrapolation black-boxed, being duly aware of the complexities inside such calculations. Note, nevertheless, his material equipment: a PC with spreadsheeting capabilities (we get to spreadsheets later), a basic calculator, larger than my hand, and paper sheets that he used for pen&paper calculations. So, for calculating flight distances, Nick was luckily equipped with an already complete fact – no need to extrapolate towards the total costs at subsidiary level: at the end of his phone call, thus, he laughed and smiled. 4 Table 1: Organised Qualculative Statements (Source: Lippert 2013, 95) However, despite the happy moment, this complete fact did not suffice on its own. It was not so complete, after all. Knowing that the organisation had paid 168,078 EUR for domestic flights did not tell him how many kilometres have been bridged. Nick reconstructed the fact corresponding to a particular mathematical form: it was the sum of several individual flights. This reconceptualisation expanded the possibility for calculability. He could now make assumptions about these individual flights, specifying these flights abstractly in two dimensions – in terms of estimated average values: one for the cost and one of the distance of an individual flight. For instance: ‘I assume, on average a domestic flight bridges a distance of 500 kilometres.’ And, he knew what he was searching for: the total distance travelled. Thus, at the outset, here, we can identify four statements (one given fact, two assumptions and one searched for result). He drew all these statements together. With Callon and Law (2005, 719) we can think of his practices as sorting out and detaching these statements from the wide range of data held by financial accountants and of the possible assumptions he could have made. They call this process qualculation, underlining the involvement of judgement and qualification with quantification and calculation; resonating with Cochoy (2008) who coined the notion qualculation in 2002. The qualification here consists precisely in performing these statements as relevant rather than different ones. Not only could have the assumptions been different (such as specifying the average distance or prices with alternative quantifiers) but also could have the statement structure been alternatively configured. The selection, thus, of precisely these statements constituted the creation and prefiguration of a space in which the subsequent calculation had to take place. Second, the space of these four statements needed to be transformed with two sets of qualities as reference points: first, the space needed to allow for calculation and, second, the calculation’s results needed to fit the material structure of the company’s environmental database, i. e. its data entry form. This form required the flight to be represented with specific qualifiers and quantifiers: these were, inter alia, first, the quantitative ‘value’ of the flight (i. e. the number) and the ‘unit’ (kilometres), second a ‘value’ and a ‘unit’ of corresponding costs and, also, a qualitative ‘comment’ on this particular data set.5 A form that satisfies these two conditions (sets of qualities) is reimagining/rewriting the statements as triples. The calculation which Nick eventually performed, thus, corresponded to a three5. Lippert (2013, 81) shows another instance of the form used by the company as employed by Nick. ‘Comment’, ‘unit’ and ‘value’ are terms used by the form. 5 fold structured space – depictable as a table (illustrated by Table 1). This space is marked with a boundary of in/exclusion: to be excluded were the traces of the statements’ origins. The fact that some statements were assumptions got lost in the framing. With this second moment Nick achieved new entities: in the table, flights and prices existed in the form of numbers, units and categories. Throughout this paper, each of these forms emerges as problematic. For now we need to stress that the notion ‘moment’, here, does not refer to a limited expansion in time and space. Using Callon and Law’s (2005, 720) voice(s) is apt. ‘What is important is that we make sure that qualculation does not get itself attached to a narrow understanding of space–time framing.’ The qualities of framing are extended from the design of the form to Nick’s preparation of data for that form – where the design and Nick’s data creation were thousands of miles and many years apart from each other. Thus, the entities drawn together by Nick were transformed and changed shape; the resulting objects did not preexist their framing. Those qualities that did not fit in, are the overflows that Callon (1998) addresses: making things calculable means framing them and that necessarily implies that some things do not fit the frame and flow over the frame – they are externalised, as the economists would have it. Here we find, thus, a form of internal externality: the project of internalising environmental consumption facts folds into itself the externalisation of the statuses of these particular environments (cf. Strathern 2005). Or, again Callon and Law (2005, 722): A boundary, so to speak, implies the cross-boundary work of making that boundary and putting things on the other side which then ‘overflow’ the division. The normal implies the abnormal. The garden implies the weeds. The boundary-drawing performs a qualification of how numbers are present in this space. Nick’s employment of the triples (in the equation) shows that he is precisely not just dealing with numbers but with interwoven qualities, categories and units. The triples do not prepare some general calculation but a much more specific qualculation. This selection, positioning and framing allows for a third moment: to actually informationally treat the four statements in relation to each other, i. e. to relate them. Nick related the costs of all flights to the average cost of a single flight. This means he treated these kinds of costs as being of a similar quality. And he handeled the average distance of a single flight as qualitatively non-distinct from the sought for total distance travelled. In both relations, he treated the two statements related as not qualitatively different but quantitatively. This treatment is called commensuration. ‘Commensuration transforms qualities into quantities, difference into magnitude.’ (Espeland and Stevens 1998, 316) That these statements are commensurable was a key judgement or qualification that Nick made. Without structuring them in this way the information would have been different. Recall, before we transformed the statements into the table, some of the statements were qualified as an assumption. The commensurability depends on the act of purifying the statements (in the second moment), transforming them into differentially treatable but directly related elements. Finally, then, the practice of relating these three triples produced a fourth triple as a result: the value with a unit for the the category of short-haul flights. To summarise, Nick related the triples that he heaved into this space in a way that allowed him to produce a result for the unknown slot. In short, with this calculation he produced the fact, the very data which he was to report. Nick, thus reported through the database environmental data that he first had to bring into existence in a thoroughly qualculative performance. 6 By way of rethinking Nick’s work as qualculating, we overcome the misleading dichotomy between calculation and judgement. In actual practice the latter overlap. Further, this take helps to identify the entities employed and makes us note that also other variables could have been used. However, here is a limit of qualculation studies: units, numbers, categories and rules of how they are related stay untouched. Our next step, correspondingly, is to focus on these entities. Doing indeterminate numbers What is the status of such qualculative entities? I reconstruct ‘the total distance of shorthaul flights’. For Nick it did not matter whether whether the sign was on paper or in a digital file. Here I recognise three entities: what the sign was for Nick, the sign itself and what the sign referred to. This is a dryadic conception, employing Charles Sanders Peirce’s dryadic semiotics in resonance with a recent STS interest in this author (cf. e. g. Muniesa 2007; Verran 2010a, 2012a, 2012b). Peirce’s approach diverges from de Saussure’s ([1916] 2011) dualistic conception of semiotics with a ‘signifier’ and a ‘signified’. Following Peirce (1940), a sign can never completely refer to an object but only partially, on a particular ground that the sign shares as a reference idea with the object. The relations between these entities are, hence, not fixed but are enacted. Suchman (2007, 77, 87) translates this into STS: signs are co-constituting situations; how signs relate to other entities cannot be found in some rule but only in the actual procedures of action. To analytically differentiate possible ways how signs can work, he proposes a three dimensional conceptual apparatus (the ‘three trichotomies of signs’); best known of which is the dimension that differentiates icons, indexes and symbols. This dimension refers to different ways how signs can refer to objects.6 Central to Nick’s work was the data entry form. The company called this form ‘task’. This designation was chosen well: the prime task of Nick was to fill the form. The task was the task itself. The name captures the partially self-referential character of the work he was to do. This understanding deserves being spelt out. His practice of assembling data was oriented towards entering it in the form. The form is also a material sign that relates to the object ‘flight’. Materially, the form shapes what a flight is, i. e. how it is to be accounted for (Robson 1992). Further, it can also be understood as an effect of all the entities and relations upon which the existence of the task-form is premised (e. g. the database programmers, the agents who configure the database, the hardware infrastructure (see Lippert 2013)). According to the work prescriptions, the distance data which Nick was to enter was to be related to an actual reality ‘out there’. This should be understood as a non-nonsensical realist commitment of the very idea of environmental accounting. Here the fact matters that flights actually are environmentally detrimental, that they cause real problems. Rather than offering mere guesswork, carbon accounting seeks to measure the emissions caused by, e. g., flights. Here, then, is a metrological enterprise that Nick was to put into practice: he was to mobilise and access signs that are truly related to the real flights. For that he was equipped with prescriptive devices that were to guide his access to reality (with e. g. a definition of which reality was to be accounted for (as part of the data entry form) as well as a paper copy of guidelines of how to enter data into the database). These prescriptions underline that the 6. The other two trichotomies differentiate the ontology of the sign itself (qualisign, sinsign, legisign) and the ways the sign relates to the interpretant (rheme, dicent sign, argument). 7 sign-to-be ‘total short-haul flight distance’ was a symbol. For Peirce (1940) this means that the sign is positioned in a prescriptive or habitual relation to the object it refers to. Verran (2010a, 2012a) proposes that in scientific accounts numbers are often presented as symbols, i. e. as determinate measures of clearly defined entities. This corresponds with the emphasis in our setting on determining the total flight distance. This works a one/many relationship in which each fragment of distance exists in a quantitative relationship with all the other fragments of distance – all replicating the same quality. Short-haul flight distance, here, has a general nature. This general nature is agreed upon – thus, the generality itself has a contractual character (Peirce 1940, 113–14). The symbol refers to things thrown together (113). Primarily, the sign itself has to be understood as a law which stipulates how the sign is to be replicated. Peirce (1940, 102) refers to the sign in its law like quality as legisign. All this resonates with the character of environmental discourse: emissions are symbols, ‘making things the same’ (MacKenzie 2009). Practically, of course, symbols are an extreme version of a sign. How signs relate to objects is always a matter of situational enactment (Suchman 2007). Analytically, the range of actual and possible ways of how signs relate to objects can be understood as an indexical space. What is indexed by a sign may be prescribed by laws but what the sign actually indexes is a matter of practice. The replica is a new configuration (272–73). At the other end of this analytical space Peirce positions icons. The total distance of flights needs also to be understood as such: it has an iconic character. As icon, total flight distance refers to the object total flight distance without the necessity for the object to actually exist; similar to ‘a lead-pencil streak as representing a geometrical line’ (Peirce 1940, 104), or an algebraic formula (105). The total flight distance, even if its formula would simply take the form of a sum rather than the form that Nick had enacted, in this sense, does not relate to any actually existing thing ‘out there’ – it is an idea. The sign total flight distance belongs to ‘a family of iconic numbers’ (Verran 2012a, 113–17). Iconic numbers bring into existence an imaginary whole; like the gross national product (GNP) or global emissions so does total flight distance. With reference to its object, the icon significantly differs from the symbol. For total flight distances to exist as iconic no actual total flight distance needs to exist. In contrast, for distances as a symbol the referent needs to exist. Interestingly, icon and symbol also overlap: both share the sign itself ‘total distance of short-haul flights’. This sign exists as a legisign, i. e. as a law of how the sign itself is to be replicated. Auditing is the corresponding societal attempts to govern that the replicates coincide in the ‘right’ ways. Both forms differ, however, with respect to how their own quality matters. Only for the sign as an icon it matters that it is text. The icon sign in relation to itself can also be understood as a qualisign. Peirce’s (1940, 101) notion is generative for analysing the total flight distance because it points us to the sign’s textual quality. And this quality effects the self-generating character of the icon: insofar as it reproduces text it stays true to its nature as text. With this discussion, however, I do not want to suggest that Nick’s practices have been treating the total distance of flights as merely or completely iconic. In fact, later the same day Nick received an email that contained a spreadsheet. The latter detailed which cost items have been part of the account for domestic flights. The list included diverse items such as restaurant visits, trips by boat and taxi or visa fees (Lippert 2013, 99–104). He considered these items out of place. In his understanding they should not have been part of constituting the flight distance fact. A boat trip did not qualify as an actual flight. Correspondingly he 8 purified the fact, got rid off its relation to things that should have not been thrown into the object. A couple of days later, the flight distance had decreased to 60 % of the distance he had calculated earlier. This discussion suggests two considerations. First, the flight distance fact was and could have been a variety of things; it could have worked as more or less indexical to the actual flights that have been employed by the subsidiary in the prior year; in theory, it could have been completely iconic or completely symbolic. Second, the fact does not have to be considered as existing in any of those options in isolation. The total distance fact can exist in several modes – as or between icon and symbol. This understanding resonates with Muniesa’s (2007) account of prices (conceptualisable as regulated law-like representations of things out there or as icons) and with Maurer’s (2008) discussion of social finance, such as islamic mortgages (which, simultaneously, do not and do contain capitalist interest fees). Verran (2012a) problematises such parallel modes of existence and meanings for the case of Costanza et al.’s (1997) ‘value of ecosystem services’ – but in this process makes present precisely the possibility of signs to not clearly refer to (any)things. Accordingly, I do not interpret Nick’s practices as effecting total flight distance as either symbolic, iconic or indexical. To propose that Nick actually did any of these would seem a philosophical fallacy. Still, the Peircian apparatus helps us differentiating the relational effects of his practices on the very ways of how the entity he related to could exist. In short: Nick did not choose any particular version of these analytical concepts. Rather, he practiced number. Therefore, next we need to scrutinise how he achieved numbers with what effects. How numbers come to matter How Nick’s practices mattered needs to be understood in relation to how further practices enrolled the entities he enacted. Therefore, briefly, I explicate the position of his practices in the company’s enactment of carbon emissions. Then I look more closely at the immediate effects of his work. I use this analysis to spell out some of the distinctions between Verran’s (2001, 2010b) account of doing numbers and Callon and Law’s (2005) of qualculation. The environmental data Nick produced was combined with data from many other subsidiaries in a central database. Headquarter-based environmental accountants studied submitted data and checked it for ‘errors’ and ‘plausibility’. This resulted in a wide variety of calls for corrections, data adjustment, each year. And over the course of several years, the central accountants learned which categories (cor)related with data entry problems. Correspondingly, not only were numbers and units adapted but also the categories themselves. In this space of data reality, modifications constituted a technically and organisationally necessarily ongoing practice. For particular reports, however, snapshots of that dataspace were transformed into more stable copies. Such were translated into reports to global ranking organisations producing environmental and economic indices; but they were also translated into glossy brochures to celebrate the corporation’s ‘sustainability’. In many of these nth -order accounts, environmental impacts existed as carbon emissions – staged as good facts, i. e. as straight-forward, audited and sufficiently objective & neutral. To achieve this translation, the central database was equipped with a differentiated and configurable conversion factor table. Technically, this tabular design allowed to translate the data produced by Nick and his global colleagues into about anything – such as water footprints or demand for ecosystem services. Therefore, these conversion factors constituted a significant ongoing presence that made data matter 9 discursively and economically – and that stabilised data into how environments exist for CEOs, policy-makers in governments and civilised NGOs7 , that is, as text. Nick’s domestic flight data, thus, needed to be translated into a form that would make his data matter. Observing the material-semiotic relations woven together between Nick, domestic flights and the database an interesting shift comes into view: while he had collected data from the financial accountants’ domestic flights file, in the environmental database he inscribed that data into the account for short-haul flights. This shift mattered in several ways. First, without the shift, his subsidiary would not have emitted any short-haul flight related emissions. Second, the domestically flown distances were mapped onto the higher conversion factor for short-haul flights; and, hence, this assemblage effected all international flights to be accounted for as long-haul flights, which were translated with a lower conversion factor. Compare this with the prescribed and assumed reality by and of the headquarter accounting system. Following the prescriptions would have resulted in mapping flights that spanned distances up to 500 km onto the category short-haul flights and all flights above this threshold value as long-haul flights. All those flights within the country that spanned distances above 500 km were consequently accounted for with a higher conversion factor than prescribed; and international flights crossing distances smaller than the threshold were translated with too low a factor. All these non-fitting lengths were partially overflowing this situated qualculative reality. In short, thus, the mattering of flights was not homogenous at all: some distances came to matter according to prescriptions, some distance fragments came to matter according to the ‘wrong’ prescriptions and, finally, some fragments of distance were not translated as mattering at all. Not only data construction with pen, paper and calculator but also data entry, thus, shapes how environment comes to matter. The actual encounter of the data with the conversion factors themselves was out of hand for Nick. Data were translated into emissions in semi-automatised reporting routines within the database. When a database user initiated an emission report, all the data for the organisation were translated via the conversion factor table. In principle, the company copied into the database the conversion factors from an industry-specific standard (see Lippert 2013, 101). Publicly, too, the company committed to this specific standard. However, for flight conversion factors the company enrolled diverging factors authored by one of the businessled ‘environmental’ organisations, the World Business Council of Sustainable Development. Their short-haul emission conversion factor was 45 % smaller than the industry specific factor. This was possible because this ‘environmental’ organisation excluded airport infrastructure from mattering into the emissions. Thus, the matter of those emissions caused by flights was not determined by any sole agent or standard. Mattering is a collective achievement. A variety of agencies were involved in shaping what resulted as emission matter: the agencies of all the company’s bookkeepers (like Nick), of the database’s algorithms as well as of the authors of conversion factors and of those who translated the factors into the database. Together they did the matter of emissions. And, depending on how they did data, different emissions resulted. Their collective doing, thus, had ontic effects – effects on what was and came to be (Verran 2010b). Matter-ing, thus, is not something ontically and normatively given but rather it is a question of concern, of value, of assembling and process, of bringing into reality, i. e. enactment (Law 2004). 7. With civilised NGOs I refer to those parts of ‘civil society’ that discuss environments at the round table rather than those uncivil collectives that actually involve in direct action, interfering in ‘out theres’ in which actual damages are effected (such as peacefully preventing airport operations). 10 It is with respect to the mode of enactment that I identify a significant difference between Verran’s (2001, 2010b) account of doing numbers and Callon and Law’s (2005) ‘qualculation’. Similarly, both locate the doing of numbers and the construction of spaces for qualculation or nonqualculation in material-semiotic generative and performative assemblages. So far I agree. Still, I see a disjuncture between these authors. To explain this disjuncture I have to get back to the notion ‘ontic’, and differentiate it from ontology. To differentiate ontic from ontological I follow Verran (2010b). Some of the agents in the company I studied also engaged in explicit reflections and deliberation upon how they constituted emissions. This could be called ontological practice – with its corresponding ontological politics. Nick’s work I described above, however, was not directly involving such ont-ology – he ‘simply’ did a reality in a particular way, doing ontics. This, too, is highly political (106–13).8 For this paper’s analysis the distinction is generative insofar as it helps to recognise members’ practices that are not necessarily actively theorised (onto-logically) but routinely or habitually enacted. Verran (2001, 159) uses the notion microworld to refer to ‘specific materially arranged times/places where rituals, repeated routine performances, occur’. She locates generalisations that bring into reality practical foundations, e. g. things, types, modes, in such microworlds. The mode of enactment here does not require the performers to possess, utilise or follow any particular ontology, a particular philosophical foundation. Much rather, members learn such rituals with the effect of these becoming self-evident. And in these ordered/ordering microworlds members do ontics routinely, and I would add, habitually. This understanding shares with Callon and Law’s (2005) the recognition of performativity. However, the mode in which the latter talk about the enactment of (non-)qualculative spaces differs. For them, such spaces are the result of qualculative practices themselves. Thus, in order to bring about spaces of (non-)qualculability, members have to qualculate: this is ‘all about arraying and manipulating entities in a space in order to achieve an outcome, a conclusion’, a result (719, my emphasis). The notion qualculation, therewith, refers to spaces of goal-oriented rationality, which, are inextricably bound to overflows. Actors know what they are doing, what they pursue. Cochoy’s (2008) discussion of shopping carts underlines the point: qualculation is about rational decision-making. All this resonates with explicitly framed spaces, spaces that are enacted based on particular ontologies. They find that qualculabilities and nonqualculabilities go together, depend on one another (Callon and Law 2005, 731), are actually interwoven (728); yet ‘the issue is whether, overall, one is left with a context or a set of events that can be qualculated or not. Or whether qualculability is pushed beyond reach.’ This requires identifying members’ and societal strategies against or supporting qualculability (726). I read this approach as heavily grounded in the Machiavellian actor. . . a story of strategic alliances, enrollment, framing and, recently, even extending to care. Both, spaces for qualculation and nonqulculation have to be qualculatingly stabilised for them to work, never perfectly succeeding. Callon and Law (2005, 730) attempt to leave behind a discussion of rationality but are strongly drawing on goal-oriented action. All this fits ontological politics. Here are actants who/which follow strategies that are designed, 8. I read Mol’s (1999) notion of ontology as not drawing this distinction. And, for studying ontological practice that directly translates into doing ontics, this distinction seems merely analytical and, therefore, not necessary. In earlier accounts of this study I do not bring the distinctions to the front (Lippert 2012a, 2013) – because I partly do find members involved in explicit reflections and deliberation upon how they constituted environmental data. 11 who/which attempt to inscribe their ontologies into the world – maybe never achieving closure but, still, struggling over the substances that may or may not be qualculable. Quite silent in the qualculation take seems to be a concern for non-strategic, non-coherent, non-aligned practices, concerns that do not matter, and whether anybody actually cares – but practices still may have ontic effects. I introduce a last empirical moment from Nick’s world. He had entered the data together with a comment that described some constituents of the calculation that produced the data. At a later point I joined a meeting between Nick and his boss. The boss demanded Nick to only report facts, no insecure estimations. Vis-à-vis the data reader, the comment in the data form, however, modalised data as not ‘perfectly’ factual (lacking objectivity, unconstructed facts of nature), it pointed the reader’s attention to the contingent nature of the fact. A few days later the comment was gone and the problem of the impression of contingency was silenced (Lippert 2013, 110–11). What did deleting the comment do to emissions? A (not quite contrafactual9 ) summary of how comments influence data processing seems apt. A data set that came with a comment was signalled to the database user at the headquarters and prompted them to review that data set. This algorithmic function served to support the system in achieving accountability, i. e. qualculability. This was considered necessary by the system designers because it was recognised that numbers not always told the stories on their own, but required some explanation (e. g. of shifts in the numbers compared to an earlier year). At the same time, however, some headquarters members expressed being happier without comments – less ‘qualitative’ comments meant less complicated data (Lippert 2011). Without comments, data users who had no direct contact to the data entry agent had a worse chance to actually learn about some of the considerations around the numbers reported. Simultaneously, no comments also meant that superiors were less likely getting back to the bookkeeper to inquire about the data reported. Both effects altered qualculability – however, in different ways. I understand the information reportable in comments as a partial account of the modality of the numbers and units reported in a form. Such modality was coconstitutive of the numbers. Bookkeepers were responsible for the data they entered. For the bookkeeper, deleting the comment also implied that they alone carried the possibility for responding to questions on data. Providing a comment extended responsibility materially to the database. The data set could respond directly to questions. A respons-able data set was also a risk – because it could answer to questions without the bookkeepers’ control. Deleting a comment made the data less accountability and reciprocally increased the qualculability of the work situation for the worker. No risk that the data set would speak against the bookkeeper. Altering a number’s modality also altered qualculability. With less qualitative comments, operations staged as purely quantitative (i. e. by way of enrolling numbers as non-qualitative, non-problematic entities) became easier, thus improving calculability. Yet, actually engaging environmental data became more problematic because the accounts were less full. With Callon and Law (2005) we can understand those headquarter voices that wanted to get rid of comments as fighting the proliferation of qualculabilities within the company. This discussion indexes a complicated space of responsibilities and accountabilities. One could ask: if Nick was not concerned about sharing the modality of data, why submit data at all? I propose that the data submission mattered in multiple ways. Nick had to submit data. For that the quality of that data did not matter too much. Nick wanted to give a good 9. I studied the role of data points with actual comments elsewhere (Lippert 2013, Sect. 4.4, 5.1, 5.2). 12 (enough) account of the flight. For that he edited the data, for indexing his data sources ‘better’. He also wanted to complete the data submission without problems (and his boss had pointed him to the friction caused by qualifying data as estimations). Correspondingly, deleting modalities became a solution. These three ways were not overly coherently aligned to one another. None of the actants involved dominated the relations around the data submission with a singular strategy. Much rather, this work space needs to be understood as ordering in multiple ways – as multiple, interfering microworlds that were both materiallysemiotically ordered and in which actants enacted a non-deterministic order of the flight fact. Precisely because the comment got deleted the resulting data was interpretable in more ways; the flight fact became more iconic, less explicitly indexing particularities in the world. The matter of the reported total short-haul flight distance of Nick’s subsidiary (in the year 2008), thus, was not precise and stable. By way of staging the flight fact as simple his ontic practice effected flight-matter with less stable meanings and as a not definitively fixed reference point. The simplicity allowed more readings of the sign. Thus, the configuration of bookkeepers, the central database and headquarter staff achieved a world of flight matter that was loose, connectable to all kinds of other entities – and not explicitly referring to the multiple microworlds in which practices and entities were ordered and performing order. In short, the distance’s matter was a relational effect, in an ANT as well as in a Peircian sense. Neither the relations themselves nor the configuration of actants were intrinsically stable. Certainties enacted Attending to how a result is qualculated analytically prefigures the achievement of closure. Accordingly when reconstructing Nick’s work, we found that he managed to relate particular qualities and quantities in a way that could produce a clearcut and simple outcome. However, when we rethink the sign produced we find that the sign in itself does not have a stable effect – it is indeterminate. Further, with understanding the matter of flight distance as an effect of the relations of how it is read and with what other information it is related, as well as of how it is algorithmically processed the matter emerges as unstable and non-singular. If environmental data is not clearly given, stable and singular, it becomes relevant to rethink how to conceptualise possibilities for certainty. My analysis yields four ways in which the qualculative practices described enact certainties: the certainty of a) a particular fact, of b) the existence of the category of that fact, of c) the practical achievement of a fact and of d) the calculability and of the simultaneous practicability of the infrastructure for achieving the environmental fact. First, users of the fact can relate to the particular fact as a representation of some outthere. Engaging the fact as a symbol, the user can check whether the fact-production process was sufficiently following the rules of how it was to be produced. The user, then, can take the number-fact, 365,387 km, as data, as given. The fact became certain. Second, additional certainties achieved are the existence of a total flight distance as well of units like length. Both are icons. They exist on paper, in the database – quite independently of whether actually something like a kilometer or a total flight distance actually exists. The total distance of short-haul flight emerged here as just another iconic number. Independently of the particular quantity, the relations achieved bringing into being and 13 stabilising a particular whole, the category of flight distances.10 This interpretation differs from Verran’s (2012a, 116) proposal that iconic numbers have to have specific referents and from her consequential critique of Costanza et al.’s (1997) number. Verran problematises the latter number for not specifically referring to particular referents. She juxtaposes this number with good iconic numbers that, in her view, seem to succeed in referring specifically. An example for her is Stern’s (2006) numbers on costs of climate change. I doubt that this juxtaposition is helpful for understanding how iconic numbers produce certainty. Stern’s (2006), Costanza et al.’s (1997) and Nick’s numbers: all of them achieve to stabilise a particular whole; all of them relate in problematic ways to (non)specificities, all miss stable reference relations. Nick’s achievement of total flight distance is a great case to show how certainty that such total flight distance exists is achieved. The certainty that this icon refers to something specific is enacted by way of postulating it and making it refer to other, already accepted, signs that are imagined as in-principle practically achievable. That something is practically achievable creates a third type of possibility for certainty. Nick managed to fill the form. This meant he completed his task – job done. He could be certain about this. This understanding is the outcome of conceptionalising him as an actant who assembled various other actants and manages their relations. It is such work that established his work as an achievement. Thus, the sign total distance flight can also be understood in relation to the user (resonating with the agent of Peirce’s interpretant). When Nick had received the disturbing spreadsheet, the distance fact became destabilised – it appeared suddenly polluted. In Peircian terms, we can understand this as a collapse of a dicent sign, i. e. of a sign that appears in its interpretations as an index an object. It sufficed for his job to purify the sign he had already materialised. It was necessary to open up the implications of the ‘logical’ impurity (such as asking whether flights may have been paid through other accounts than the subsidiary’s travel account). He, eventually, had not only invested his qualculative competency to produce a sign that fitted the form but also two types of purification work: purifying the numbers from visa costs and restaurant visits as well as purifying the data entry form from a comment that indexed some of the details of how the numbers have been qualculated. Having worked across all these ‘layers’ the data entry form, eventually, was defined as completed with a tick-box in the form itself. This created the certainty that the practical work on this fact was sufficiently completed. Finally, I propose, certainty of calculability is maintained. That flights and other environmentally relevant processes and their impacts are accountable in practices that are imagined as being a matter of numbers is not self-evident. If, e. g., auditors or public service agents assure certainty for the earlier forms of certainty then certainty is achieved not only with respect to the signs of environmental data but also with respect to those agents that certify the data and with respect to the rationalities and infrastructures these agents co-enact. Imagining the possibility of calculability underlies a variety of highly relevant societal relations: I see number making as part of extending the realm of the performance of manageability and governability. That companies are manageable, that environments and companies’ impacts upon them are governable is not a given but requires precisely such work as Nick has achieved. His practices were part of co-performing the practico-material-semiotic foundation of a range of societal relations. Regimes of accounting, i. e. the imaginary that accounting actually works well, that accounting rules are apt (Porter 1992), are maintained – with 10. For this argument I have to credit conversations on Verran’s (2012a) paper with Niklas Hartmann. 14 the effect of producing comforts amongst professional accountants, auditors and consultants (Pentland 1993). The universal applicability of a mathematical apparatus, of ‘[q]uantifying [as] thoroughly stabilized routines’ (Verran 1999, 150) is reproduced, and the apparatus itself is re-enacted (da Costa Marques 2004). We can even see this practice as providing the foundation for what Porter (1995) calls Trust in Numbers and Power (1999) The Audit Society. Others entertain optimistic considerations: numbers as opening up spaces for political contestation for these numbers allegedly provide a shared and clear-cut point of reference for deliberation (the ‘old sociological story of rationalisation’, Barry 2005, 88), or at least as co-constituting metrological regimes that co-configure political realities in non-deterministic ways. All this is grounded in the certainty that calculability is possible. If Nick and all those agents of modernisation did not conjure up a stage of certainty, the story that calculability is certain would appear immediately at high risk: too much evidence signals that regimes of calculability ‘fail’ (e. g. Scott 1998; Gilbertson and Reyes 2009). These kinds of certainties resonate with Peirce’s conception of a sign through how it can exist in relation to itself, to its object and to the interpretant of the sign. I found certainties at stake in relation to the object of the sign (certainty of a particular fact), to the sign itself (the existence of the category of the fact), to the practical achievement of the sign by the interpretant. In addition to this triadic approach I found that the possibility of calculability is at stake (maybe this resonates with the possibility of language). Nick was involved with a dynamic space of data enactment. This space was not deterministic at all. However, this did not mean it was arbitrary, completely fluid or unstructured. The space can be better conceptualised in terms of landscape, effecting a datascape or, for carbon-constituted-through-data-practices, a carbonscape (Lippert 2013, Ch. 6). Nick’s practices helped maintaining this carbonscape. The maintenance of the possibility of this landscape was key: entering data in the database (in the manner he did) was perfectly contributing to performing the ground of the company’s emission facts. It allowed easily including ‘his’ facts in nth -order qualculations, it allowed for parallel forms of certainty. In general, we learn here, several forms of certainties are possible. The more forms of certainties are made practically and societally relevant (through additional forms of quality/quantity assurance regimes), the more chances companies and workers have to achieve at least some form of certainty. Insofar, however, as certainty is conceptionalised in totality (i. e. as some thing is certain or it is not), the staging of total certainty is prefigured: a significant analytical risk seems to be to conflate these various forms of certainty. If we conflate them we may effect ‘collateral’ certainties (Law 2012). It seems advisable, then, to always question which certainties we are buying into, co- and reproducing. Conclusion Environmental and much other debate is increasingly mediated and performed in quantitative terms. Numbers, hence, figure prime in corresponding discourse. This raises the question: how do numbers’ practitioners effect numbers as certain? To address this theme, I have traced the work achieved in the process of performing a real-world calculation (to be solved for one unknown). I find data is enacted. For the practical purposes of accountants, they have to mobilise data or construct that data in order to have some matter which they can formate in a way that fits a data entry form. Data is not a simple given but is a practical, material and 15 semiotic achievement. Data takes place – in a world of words, numbers, forms, boundaries, adjustments, purifications and shifts. Quantitative data is an abstraction; in practice numbers exist always materially (e. g. on a screen, on paper, in voice, etc); numbers exist only in relation to qualifiers (such as units, categories, forms, users). Drawing on Peircian semiotics I identify numbers as unstable signs: they are constituted in ongoing practices of relations – relations between the objects they refer to, the signs themselves, the users and their interpretations. In this contingent field of possibilities of what numbers are, several forms of certainty can exist. This allows numbers’ users to easily stage their numbers in at least some sense as certain. It follows that a number appearing as a straight-forward fact may actually co-constitute an open space of certainties. Calculations enact certainties; and meta-orders of calculations, like auditing, effect further orders of certainty. The more possible certainties exist, the less we know what it is we are certain about. Ontic uncertainty results. Still, users may perform a number as certain. Like Callon and Law (2005) I find calculations do not necessarily result in closure. They differentiate qualculative from nonqualculative engagements. A case of the latter is care. The issue in care ‘is about what to do, here and now, with’ an entity that is affected by relations of suffering (728). I propose: number work involves not just qualculation and nonqualculation, i. e. not just attempts to get environments in control and attempts to actually care for environments. Rather, doing numbers may effect a space of possible certainties, i. e. of vagueness, uncertainty. The environmental data produced, treated as numbers, is shaped in a way that destroys the relations of reference to the specific practical and environmental realities which co-constituted the data. Precisely because environmental data is translated into clearly structured signs (e. g. triples of quantifiers, units and categories) nth -order reorderings are made more easily possible. The data produced becomes co-constitutive of a space of data that allows for a wide variety of data treatments and figurations. In the company studied that datascape appears as carbonscape, a landscape of carbon emission realities that data practitioners can and do order and reorder. The less the datasets that this datascape is founded upon refer to specificities the more possible becomes qualculability, performed as calculability, and, thus, the enactment of certainty. This paper therewith provides the ground for this consideration: does the study of numbers in relation to environmental concerns specifically diverge from studies of (non)qualculation around other discourses, such as health, care, science or engineering particular technologies? I venture that in the company’s engagement (a green leader according to e. g. the Dow Jones Sustainability Index ) the enactment of environments as datascapes, of carbon emissions as carbonscape, allows for tactical data practices that underwrite performances of shifting sustainabilities while the more decoupled that datascape is from the relations of which it was produced the more flexible the tacticians can employ the datascape. Therewith, doing environmental numbers can help to effect stability for the corporation in a discursive environment that urges to revolutionise company-economy-human-environment relations. The less specific data is, the better it allows to perform tactical sustainabilities – sustainabilities that can simultaneously be performed as certain. The possibility for nonaction (on environmental issues) is, thus, explained precisely in relation to the translation of environmental concerns into calculative engagements. Calculating environments enacts certainty and trust in the company – while impossibilising both, control and care of and for environmental issues. This interpretation resonates with recent theorising in environmental 16 sociology that draws on STS by Luigi Pellizzoni (2011). He proposes that governing environments is now pursued by way of playing out uncertainty and disorder. My account grounds such theorising in a study of how certainty is enacted. For the data practitioner and the corporate environmental tactician, some form of certainty is always possible. If you get the calculation ‘right’, the least you can do is to claim some form of mathematical certainty. STS scholars have focussed very much on practitioners who are committed to control or care. This paper scrutinised relations that perform control but systematically order microworlds in a way that render both, care and control, difficult. Even though economics may claim that markets demand certainty of the goods and services traded, such certainty may actually be quite partial. This work, thus, extends Callon et al.’s (2005) take of qualification of goods and services: the latter may be qualified even in cooperation with consumers, resulting in a wonderful state in which consumers get exactly what they want (38, 44). Certainly, I propose, consumers may be presented with what they want – but such statement does not sufficiently qualify the presence of the things produced. I find a company can easily offer consumers precisely those types of sustainabilities they wish to see: simply reorder the datascapes accordingly. The data they are grounded in have been carefully stripped off significant references that would constrain corporate performances of themselves as green. These considerations strongly challenge the optimism in the work of scholars, e. g. Thrift (2004, 592), who consider the new worlds of qualculations as protheses for humans – so they could act more freely in societies and natures. Instead, with the proliferation of performances of qualculations and simultaneous of certainties I see the impossibilisation of direct responsabilities. Translating environmental problems into questions of accounting (and based on that into markets) produces what Beck (1988) calls organised irresponsibility. To some degree, then, I can follow Moser and Law (2006, 68) who postulate more ‘permissive strategies for informatisation’ that should allow for ‘multiple modes of qualculation’. Well, yes, data work should be organised to allow the latter – but not at the cost of conflating the resulting multiple respons-abilities with citizen or consumer control, let alone care, over the entities that are supposedly represented in datascapes. Rather, if we search for better informed ways of how humans can relate to natures, then it would be relevant to allow affected humans to directly oversee how natures are formatted and shaped in the process of enacting data. References Barry, Andrew. 2005. ‘The anti-political economy’. In Barry and Slater 2005, 84–100. Barry, Andrew, and Don Slater, eds. 2005. The technological economy. London and New York: Taylor & Francis. Beck, Ulrich. 1988. Gegengifte: die organisierte Unverantwortlichkeit. edition suhrkamp. Frankfurt am Main: Suhrkamp. Callon, Michel. 1998. ‘An essay on framing and overflowing: economic externalities revisited by sociology’. In The Laws of the Markets, edited by Michel Callon, 244–69. Oxford, Malden: Blackwell. . 2009. ‘Civilizing markets: Carbon trading between in vitro and in vivo experiments’. Accounting, Organizations and Society 34, nos. 3-4 (April): 535–48. doi:10.1016/j.aos.2008.04.003. 17 Callon, Michel, and John Law. 2005. ‘On qualculation, agency, and otherness’. Environment and Planning D: Society and Space 23, no. 5 (January): 717–33. doi:10.1068/d343t. Callon, Michel, Cécile Méadel and Vololona Rabeharisoa. 2005. ‘The economy of qualities’. In Barry and Slater 2005, 28–50. Cochoy, Franck. 2008. ‘Calculation, qualculation, calqulation: shopping cart arithmetic, equipped cognition and the clustered consumer’. Marketing Theory 8 (1): 15. doi:10.1177/1470593107086483. Costanza, Robert, Ralph d’Arge, Rudolf de Groot, Stephen Farber, Monica Grasso, Bruce Hannon, Karin Limburg, Shahid Naeem, Robert V. O’Neill, Jose Paruelo, Robert G. Raskin, Paul Sutton and Marjan van den Belt. 1997. ‘The value of the world’s ecosystem services and natural capital’. Nature 387 (6630): 253–60. doi:10.1038/387253a0. da Costa Marques, Ivan. 2004. ‘Mathematical metaphors and politics of presence/absence’. Environment and Planning D: Society and Space 22 (1): 71–82. doi:10.1068/d319t. de Saussure, Ferdinand. (1916) 2011. Course in general linguistics. Edited by Charles Bally, Albert Sechehaye and Albert Rledlinger. New York, Toronto, London: McGraw-Hill Book Company. Espeland, W.N., and M.L. Stevens. 1998. ‘Commensuration as a Social Process’. Annual Review of Sociology 24, no. 1 (August): 313–43. doi:10.1146/annurev.soc.24.1.313. Gilbertson, Tamra, and Oscar Reyes. 2009. ‘Carbon Trading: How it works and why it fails’. Critical Currents, no. 7 (7th September). Accessed 15th June 2013. http://www.dhf.uu.se/pdffiler/cc7/cc7_web.pdf. Grove, Richard H. 1996. Green imperialism: Colonial Expansion, Tropical Island Edens and the Origins of Environmentalism, 1600-1860. Studies in Environment and History. Cambridge: Cambridge University Press. Lave, Jean. 1988. Cognition in practice: Mind, mathematics and culture in everyday life. Cambridge: Cambridge University Press. Law, John. 2004. ‘Matter-ing: Or How Might STS Contribute?’ Published by the Centre for Science Studies, Lancaster University, Lancaster LA1 4YL, UK. 30th June. Accessed 27th March 2005. http://www.lancs.ac.uk/sociology/papers/law-matter-ing.pdf. . 2012. ‘Collateral realities’. Chap. 8 in The Politics of Knowledge, edited by Patrick Baert and Fernando Domı́nguez Rubio, 156–78. London and New York: Routledge. doi:10.4324/9780203877746. Law, John, and John Hassard, eds. 1999. Actor Network Theory and after. The Sociological Review. Oxford, Malden: Blackwell. Lippert, Ingmar. 2011. ‘Extended Carbon Cognition as a Machine’. Computational Culture 1 (1). http://computationalculture.net/article/extended-carbon-cognition. 18 Lippert, Ingmar. 2012a. ‘Carbon Classified? Unpacking Heterogeneous Relations Inscribed Into Corporate Carbon Emissions’. Ephemera 12 (1/2): 138–61. http://www.ephemeraweb.org/journal/12-1/12-1lippert.pdf. . 2012b. ‘Carbon Dioxide’. In Encyclopedia of Consumption and Waste: The Social Science of Garbage, edited by Carl A. Zimring and William L. Rathje, 105–8. Sage Publications, March. doi:10.4135/9781452218526.n44. . 2013. ‘Enacting Environments: An Ethnography of the Digitalisation and Naturalisation of Emissions’. PhD thesis, Augsburg University, Philosophisch-Sozialwissenschaftliche Fakultät. http://nbn-resolving.de/urn:nbn:de:bvb:384-opus4-21997. . fortcoming. ‘Studying Reconfigurations of Discourse: Tracing the Stability and Materiality of “Sustainability/Carbon”’. Journal for Discourse Studies — Zeitschrift für Diskursforschung 1 (3). Lohmann, Larry. 2009. ‘Toward a different debate in environmental accounting: The cases of carbon and cost–benefit’. Accounting, Organizations and Society 34 (April): 499–534. doi:doi:10.1016/j.aos.2008.03.002. MacKenzie, Donald. 2009. ‘Making things the same: Gases, emission rights and the politics of carbon markets’. Accounting, Organizations and Society 34, nos. 3-4 (April): 440–55. doi:10.1016/j.aos.2008.02.004. Maurer, Bill. 2008. ‘Resocializing finance? Or dressing it in mufti? Calculating alternatives for cultural economies’. Journal of Cultural Economy 1 (1): 65–78. doi:10.1080/17530350801913668. Mol, Annemarie. 1999. ‘Ontological politics. A word and some questions’. In Law and Hassard 1999, 74–89. Moser, Ingunn, and John Law. 2006. ‘Fluids or flows? Information and qualculation in medical practice’. Information Technology & People 19 (1): 55–73. doi:10.1108/09593840610649961. Muniesa, Fabian. 2007. ‘Market technologies and the pragmatics of prices’. Economy and Society 36 (3): 377–95. doi:10.1080/03085140701428340. Peirce, Charles Sanders. 1940. Philosophical writings of Peirce. Edited by Justus Buchler. New York: Dover Publications. Pellizzoni, Luigi. 2011. ‘Governing through disorder: Neoliberal environmental governance and social theory’. Global Environmental Change 21:795–803. doi:10.1016/j.gloenvcha.2011.03.014. Pentland, Brian T. 1993. ‘Getting comfortable with the numbers: auditing and the micro-production of macro-order’. Accounting, Organizations and Society 18 (7-8): 605–20. doi:10.1016/0361-3682(93)90045-8. Porter, Theodore M. 1992. ‘Quantification and the accounting ideal in science’. Social Studies of Science 22 (4): 633–51. doi:10.1177/030631292022004004. 19 Porter, Theodore M. 1995. Trust in Numbers: The Pursuit of Objectivity in Science and Public Life. Princeton, New Jersey: Princeton University Press. Power, M. 1999. The Audit Society: Rituals of Verification. Oxford: Oxford University Press. Robson, Keith. 1992. ‘Accounting numbers as “inscription”: Action at a distance and the development of accounting’. Accounting, Organizations and Society 17 (7): 685–708. doi:10.1016/0361-3682(92)90019-O. Scott, James C. 1998. Seeing like a state: How certain schemes to improve the human condition have failed. The Yale ISPS Series. New Haven: Yale University Press. Sørensen, Estrid. 2012. ‘The mind and distributed cognition: The place of knowing in a maths class’. Theory & Psychology 22 (6): 717–37. doi:10.1177/0959354312436869. Stern, Nicholas Herbert. 2006. Stern Review: The economics of climate change. London: HM treasury. Strathern, Marilyn. 2005. ‘Externalities in comparative guise’. In Barry and Slater 2005, 66–83. Suchman, Lucy. 2007. Human-machine Reconfigurations: Plans and Situated Action. 2nd ed. New York: Cambridge University Press. Thrift, Nigel. 2004. ‘Movement-space: The changing domain of thinking resulting from the development of new kinds of spatial awareness’. Economy and Society 33 (4): 582–604. doi:10.1080/0308514042000285305. Verran, Helen. 1999. ‘Staying true to the laughter in Nigerian classrooms’. In Law and Hassard 1999, 136–55. . 2001. Science and an African Logic. Chicago and London: University of Chicago Press. . 2010a. ‘Number as an inventive frontier in knowing and working Australia’s water resources’. Anthropological Theory 10 (1-2): 171–78. doi:10.1177/1463499610365383. . 2010b. ‘The educational value of explicit non-coherence’. Chap. 6 in Education and technology: Critical perspectives, possible futures, edited by D. Kritt and L.T. Winegar, 101–24. Lanham, Maryland: Lexington Books. . 2012a. ‘Number’. Chap. 8 in Inventive methods: the happening of the social, edited by Celia Lury and Nina Wakeford, 110–24. London, New York: Routledge. . 2012b. ‘The changing lives of measures and values: from centre stage in the fading “disciplinary” society to pervasive background instrument in the emergent “control” society’. The Sociological Review 59:60–72. doi:10.1111/j.1467-954X.2012.02059.x. 20