publisher colophon
  • Doing Science Justice:Speculative Materialism and the Facticity of Research

The contemporary climatic situation is as urgent as it is frustrated. The pace of movement on international agreements, public opinion and national legislation is at odds with the time frames suggested by many scientific estimates. In this context it seems vital to seek new philosophical resources for responding to the sciences of the climate. Quentin Meillassoux’s philosophy is appropriate for this task for two reasons. Firstly, he believes that what he calls “speculative materialism” is a new system of thought. Its philosophical novelty may open up new possibilities for relating to science and to the environment. Secondly, the novelty of his approach emerges out of the broad narrative he winds around contemporary paradigms of thought, the conventional background against which he wants to make his mark (he is largely concerned with philosophy, but this story easily spans other humanities disciplines and the social sciences too). Just as one might assert oneself philosophically against such a framing of the status quo, so a politics of climate may be aided from better diagnosis of the contemporary situation, the obstacles and assets of the landscape through which it is to be accomplished.

I use this paper to expand Meillassoux’s argument towards its potential implications for climate science. This is a challenging task, of which I can only really suggest the beginnings and a few initial directions. But I want to argue that this kind of philosophy could hold great potential in this area, and that it will be well worthwhile those of us interested in the problems of the climate following closely its future development.

Meillassoux has emerged as a prominent figure within a younger generation of thinkers, many of whom are looking to him for signs of the future path of continental philosophy. His rise in fame has been buoyed by his association with the “speculative realism” movement (see, for example, Bryant, Srnicek, and Harman), as well as a tide of interest, over the last decade or so, in the work of his teacher, Alain Badiou. Meillassoux presents his philosophy as something of an architectonic, a grand system for thought, [End Page 163] but it is important to bear in mind that this anticipated scope reaches far beyond what has yet appeared in print (so far a handful of articles and two short books). I shall here confine myself to discussing After Finitude, where he gives the most systematic treatment of his ideas, but which, by the admission of its own subtitle, modestly proclaims itself to be only an “essay.”

As with the other essays in this special issue, the problem of the climate provides me with a central motivation and it is this I discuss first. That being said, the question I derive from this problem is one of how we relate to science in general, so I spend little time discussing the climate in particular. The second section addresses the theoretical challenge posed by the somewhat problematic position of language in Meillassoux’s relationship with science. I then spend the third section of the essay outlining the core of Meillassoux’s argument, before moving on in the final section to relate his philosophy to theories from Science and Technology Studies. I argue that the concept of “facticity” captures something of the driving force of scientific practice, and that Meillassoux’s treatment of this concept may provide a novel and informative dimension to our analysis, potentially opening new options for approaching science.

Approaching Climate

In this essay I explore our relation to the climate through the problem of “doing science justice.” It is important to note at the outset, however, that this is by no means the only manner of engagement. Most significant among alternatives would be a philosophy of the environment, often thought phenomenologically, in terms of the world as the horizon of human existence (see, for example, McWhorter and Stenstad 2009): “That world of everyday Dasein which is closest to it, is the environment” (Heidegger 1978, 94). Being-in-the-world, in the Heideggerian manner of speaking, is a fundamental attribute of Dasein, of being-there, through which it is possible to think of existence in terms of being “thrown” into a world of involvements. We are thus originally close to our environment, and a philosophy of the primordial phenomena of existence can grasp it as something with an ontological legitimacy, thus establishing a strong foothold for critique.

Such an approach, however, is somewhat awkward if our aim is to adequately grasp the specific problem of the climate. If you start from a phenomenological view of the environment, it is hard to comment on scientific abstractions. In contrast to the environment, the climate is always mediated by a number of scientific procedures (measurement, modeling, mathematization), and from a phenomenological point of view it will always seem derivative and secondary. Furthermore, it is very difficult to claim that the cultivation of a good relation to an environment would necessarily add up to a healthy climate. If climate is the primary problem, it is necessary to [End Page 164] orient ourselves towards the scientific practices through which it becomes an issue, and here Meillassoux’s philosophy could serve us well.

One of the central problems for Meillassoux is the question of what the right relationship might be between philosophy and science. In After Finitude, he claims that post-critical philosophy (i.e., modern philosophy after Kant) can be defined by a tendency to place the human relation to the world at its very centre (2009, 3-8). Science, on the other hand, works differently. It produces statements that don’t refer to that relation at all, that speak directly of the world, so that when philosophy approaches science it does so according to a logic that is at odds with that of its object. Meillassoux spells out this argument with respect to the case, strategic for his purposes, of scientific knowledge of things that existed prior to humans, paradoxical for philosophy because they are prior to any such human relation to the world: “How are we to grasp the meaning of scientific statements bearing explicitly upon a manifestation of the world that is posited as anterior to the emergence of thought and even of life – posited, that is, as anterior to every form of human relation to the world?” (2009, 9-10, emphasis in original).

A modern philosopher, when faced with a scientific statement (in this case about the accretion of the earth)

will not contest the claim that it is in fact event Y that occurred, nor will she contest the dating of this event. No—she will simply add—perhaps only to himself, but add it he will—something like a simple codicil, always the same one, which he will discretely append to the end of the phrase: event Y occurred x number of years before the emergence of humans—for humans (or even, for the human scientist).

(2009, 13; emphasis in original)

This dominant mode of thought Meillassoux calls “correlationism.” Its many forms include transcendental philosophy, phenomenology and philosophy of language. In each case, knowledge cannot take us beyond its site. What is common among them is the foregrounding of the idea of the correlation, “the idea according to which we only ever have access to the correlation between thinking and being, and never to either term considered apart from the other” (5). This extrinsic epistemological framing prevents science being taken on its own terms, a frame through which it appears “that science harbours a meaning other than the one delivered by science itself,” a great injustice (119). But before we tease out what alternative Meillassoux might offer, it is necessary to clarify some potential confusion arising from how he sets out the issue. [End Page 165]

Language and Science

Given the historical dominance of correlationist positions, it is hardly surprising that we are thrown deep into troublesome waters in the very process of trying to grasp the problem. What is this meaning that would be “delivered by science itself”? It would be all too easy to paraphrase Meillassoux and formulate the issue as one of allowing science to speak for itself. But in making this move, we already frame the question in terms of language, which in Meillassoux’s eyes forms one of the most pervasive correlational systems of the twentieth century. Language, he says, is one of the “two principle ‘media’ of the correlation” (the other being consciousness) (Meillassoux 2009, 6).

While he is a staunch critic of language-centric philosophy, Meillassoux treads a very precarious line when he introduces the problem of correlationism by taking as exemplary a scientific statement about the date of the accretion of the earth. He takes a huge risk of being misunderstood in this move. He risks his argument with the correlationist tendency becoming merely a dispute over the interpretation of statements, over whether on the one hand they should be taken literally, or on the other whether they should be taken to be relative to their site of enunciation. This is exactly what he needs to avoid, and we need to bolster his thought on this matter if we are to be generous to his project.

If all Meillassoux wanted was for the statement about the accretion of the earth to be taken literally, he need make no novel theoretical move whatsoever. While perhaps rare among continental philosophers, in a wider forum “naïve” kinds of objective realism are very widespread (for a good introduction to these themes, see Giere 2006, 4–6). It is not the slightest bit unconventional to regard scientific statements as literally true. Furthermore, the trope of scientific objectivity is historically co-extensive with the heyday of correlationism (see, for example, Daston and Galison 2007). It is not going to bring anything new into the mix to turn towards this trend.

How can we gain some perspective on this problem? How can we bolster the argument such that we retain the sense that Meillassoux is proposing something more radical than yet another philosophical sanctioning of objectivity? To answer this question it is necessary to appreciate that Meillassoux does not believe it is possible to simply wash our hands off our correlationist heritage. It is necessary to work from within it.

Meillassoux is distinctive among his peers in the speculative realism movement because he works through correlationist positions. Unlike Graham Harman, for instance, he does not believe that correlationism is simply a mistake, to be discarded (Harman 2011, 131). He believes it can and must be overcome based on its own internal logic. If this were going to be done, it would be well advised to begin with a well-founded correlationist position and work from there. [End Page 166]

In formulating his initial problem vis-à-vis science, Meillassoux would do well to learn from Andrew Pickering, who stressed in the 1990s that the major lesson of the empirical studies of scientific research that had been carried out over the previous twenty years had been the need for a decisive move “from science as knowledge to science as practice” (1992, 1). While scientific work does produce linguistic outputs, it also produces many more things. It produces refined technical systems, material traces, scientific objects and habituated bodies. Innovative research emerges from the iterative self-reproduction of these systems (Rheinberger 1997, 3). While these perspectives from Science and Technology Studies (STS) would still count as correlationism, it is a correlationism derived from empirical studies of the actual workings of scientific research, rather than from a priori assumptions about the nature of knowledge. It is thus a much more appropriate starting point for developing Meillassoux’s position. The question is: how do you do justice to these practical systems of research? How do you find the opening in which they are not merely correlational? That is the kind of question that Meillassoux might be able to answer.

Part of the reason for these preparatory difficulties is that the project is incomplete. After Finitude sets the scene, but “it is not our aim here to resolve this problem; only to try to provide a rigorous formulation of it” (2009, 26). Yet Meillassoux will hint at what will come next, and here again we must be very careful not to slip into a literalist objectivism. Meillassoux suggests that his system will eventually lead to an absolutisation of mathematics, and through it, an absolutisation of scientific discourse (126). In this respect, Meillassoux remains close to Badiou, who in Being and Event famously claimed that mathematics is the science of Being (2007, 4).

It is hard to see how a resuscitation of unificatory projects might be at all radical. However, if we read Meillassoux carefully, he may not be wholly committed to unifying science under mathematics. He talks of “re-absolutizing the scope of mathematics,” of deriving “the capacity, proper to every mathematical statement, through which the latter is capable of formulating a possibility that can be absolutized, even if only hypothetically” (2009, 126). We are talking about a possibility inherent in mathematical abstraction, a possibility that remains powerful even if it remains only a possibility. This does not imply the objectivity of the discourse of the mathematical sciences. It is a way of locating a power in formalization that implies that mathematics, as a practical operation, is more than a way of speaking. What I would want to leave open is the possibility, one which Meillassoux does not foreclose, of other such operations, a landscape of many such possibilities, sensitive to the diversity of scientific endeavors. [End Page 167]

The Speculative Materialist Argument

The core argument of After Finitude is complex, but it is worthwhile providing an abridged version here, because it is on this argument that speculative materialism succeeds or falls. The argument is a “proof, which could be called ‘indirect’ or ‘refutational,’ [which] proceeds not by deducing the principle from some other proposition—in which case it would no longer count as a principle—but by pointing out the inevitable inconsistency into which anyone contesting the truth of the principle is bound to fall” (2009, 61). I am attempting to strip it back to its basic form, avoiding tangential lines of enquiry, but at the cost of omitting much of the original sophistication, and Meillassoux’s many maneuvers around his anticipated opponents’ objections. For example, for brevity’s sake I elide the distinction between “strong” and “weak” correlationism. So while the reception of his thought can benefit from re-statement such as this, it must be acknowledged that there is no substitute for going back to the original.

To start, correlationism in Meillassoux’s narrative becomes a dominant feature in modern philosophy because any naïve realism, claiming knowledge of the world outside of the correlation, is “vitiated by the inconsistency proper to all realism—that of claiming to think what there is when one is not” (55). Through such arguments it becomes untenable to claim to have absolute knowledge of the nature of reality: thought can never escape itself, can never attain access to the thing-in-itself.

The term “correlationism” may seem far too all encompassing to give sufficient granularity for any particular critique. However, Meillassoux’s stipulation that we must not confuse correlationism with idealism does give purchase on some of the stakes of post-critical thought. In short, idealism takes the idea of the human-world relation and claims this idea to be an instance of absolute knowledge (37). It is not simply that all knowledge just happens to be correlational; for the idealist this is necessarily the case. If the correlation is necessary, any concept of a world without the correlation is meaningless. Such an outside reality is contradictory because it can only be meaningful through the correlation, and hence can never really be outside it. From a sociological point of view, in which the correlation may take a plurality of forms according to its origin in a culture, a research school or a moment in history, this idealist move makes the relative incommensurability of frameworks of knowledge an absolute fact, leading us to twin this denial of reality with something akin to cultural relativism. If the correlation is taken for an absolute, any extra-human reality is a secondary effect of particular positions, not anything with a claim to independent existence. Therefore, despite the wide scope of Meillassoux’s narrative, he nevertheless gives us a platform to think about dilemmas in science studies. We can gain some perspective on the Science Wars, because we can extricate correlationism proper from the relativisms that may be derived from it. What was so objectionable to many during these extremely divisive debates was perhaps [End Page 168] not correlationism as such, but rather the derivative idealist position, the thesis that the objects of scientific knowledge have no reality outside of the process of their construction.

Meillassoux notes that the correlationist may resist its idealist opponent by pointing out that the idealist can give no reason for this postulated absolute necessity of the correlation. While we may indeed always be bound up in a correlational view of the world, we have no access to any kind of reason why this must necessarily be so or why it must take the form it does. Just because such an outside has no meaning for us, just because we cannot step outside of our perspective on the world without abandoning sense, does not mean an external reality does not exist. The idealist is shown to rely on an equation of sense with existence, one which could only be justified if the correlational realm can be deduced, proven necessary. The correlationist’s next move is to claim that we have no access whatsoever to any reason that would ground such a deduction, and that idealism is therefore illegitimate (38).

In contesting realism and resisting idealism, the correlationist takes a stance against any claim to have access to an absolute, but rather than this being proof that an absolute does not exist (for that would be a kind of absolute knowledge), we end up with a kind of philosophical agnosticism, one which states that philosophy (and I would want to extend the claim to sibling enterprises such as sociology) simply cannot legitimately speak of such things. Where religious discourse, for example, speaks of the absolute, philosophy must remain silent (43–48).

It is here that Meillassoux intervenes to establish his own position. He will support the correlationist against both the objective realist and against the idealist, but will then attempt to push it further. His claim is that we can find a way in which correlationism does indeed yield positive knowledge of an absolute, that it cannot be as agnostic as it might wish. The correlationist opposition to absolutes is specifically an opposition to “metaphysical” absolutes, defined as the absolute necessity of some entity. Meillassoux’s strategy rests on his identifying a non-metaphysical absolute that has slipped under the correlationists’ radar, that actually may be logically implied by their very position. But what could such an absolute possibly be? Meillassoux’s answer: “The necessity of contingency.”

We derive absolute knowledge of the necessity of contingency through an examination of the correlationists’ position. Recall that the latter is to be distinguished from the idealist through the argument that although all knowledge is correlational, we have no access to a reason why this must necessarily be the case. But what, asks Meillassoux, is involved in thinking this lack of reason? It must be thinkable for the argument to work. It cannot be a specific possibility of the correlation being other than what it is, for this would fall into the contradiction of claiming to think something outside of thought. The thought on which the argument rests is rather an “open possibility, wherein no eventuality has any more reason to be realized than [End Page 169] any other” (58). It is through such an opening that the correlationist thinks the lack of necessity of the correlation, and strikes a blow at the idealist.

“Facticity” is the name given to the correlation’s lack of reason. The facticity of human existence and human relationships with the world is something well established in phenomenology, particularly in Heidegger’s Being and Time, where he considers it to be a fundamental attribute of Dasein: “Whenever Dasein is, it is as a fact; and the factuality of such a Fact is what we shall call Dasein’s ‘facticity’” (1978, 82). Likewise, for Meillassoux, facticity is more than the contingency of a particular fact. It is the lack of reason behind the very form in which facts are encountered, the form of the correlation:

[F]acticity…pertains to those structural invariants that supposedly govern the world—invariants which may differ from one variety of correlationism to another, but whose function in every case is to provide the minimal organization of representation: principles of causality, forms of perception, logical laws, etc. These structures are fixed—I never experience their variation, and in the case of logical laws, I cannot even represent to myself their modification…. But although these forms are fixed, they constitute a fact, rather than an absolute, since I cannot ground their necessity—their facticity reveals itself with the realisation that they can only be described, not founded.

(2009, 39)

Whereas the correlationist and the idealist differ on the question of whether the correlation is a fact or an absolute, Meillassoux’s innovation is to ask the correlationist whether the facticity of the correlation is a fact or an absolute. Can facticity itself be a fact? Meillassoux’s answer is no—this does not make sense. If facticity is not a fact, could it be an absolute?

Meillassoux wants to show that facticity must be an absolute by showing that the correlationist’s whole defense against idealism implicitly depended on it. The issue comes down to whether facticity is a possibility of ignorance, which stops us getting at the reason hidden behind everything, or whether it is an absolute in its own right. On the one hand, we would rely on Leibniz’s principle of sufficient reason, in terms of which facticity would either sum up our ignorance of the ultimate reasons behind things, or, where we have do knowledge, it would capture the fallibility of that knowledge and the fact that it is always possible to seek ever deeper reasons, a level of hidden necessity at which we are ignorant. On the other hand, we have Meillassoux’s “principle of unreason,” the necessity of contingency, which maintains that there is no ultimate reason, that everything is absolutely contingent: “We must grasp in facticity not the inaccessibility of the absolute but the unveiling of the in-itself and the eternal property of what is, as opposed to the mark of the perennial deficiency in the thought of what is” (52). In disqualifying a realm of hidden reasons lying beneath, Meillassoux is thus giving us a new kind of radically immanent ontology in which we “put back into the thing itself what we mistakenly took to be an incapacity in thought” (53, emphasis removed). [End Page 170]

Meillassoux’s speculative materialism emerges from the recognition that, all along, his principle (not Leibniz’s) had been at the heart of correlational systems of thought. If facticity were only an attribute of the correlation, then it would be impossible to defend correlationism against idealism. If facticity is a limitation internal to the correlation then it cannot be a possibility of this correlation being other than itself. The only alternative is therefore to accept that facticity is an absolute, albeit a non-metaphysical one, giving us absolute knowledge of the principle of unreason, which has through this procedure been elevated beyond the correlation, the absolute necessity of contingency. As “the absolute impossibility of a necessary entity” (60) all metaphysics is disqualified, as is correlationist skepticism of absolute knowledge in general:

[I]f we can discover an ontological truth hidden beneath facticity; if we can succeed in grasping why the very source which lends its power to the strategy of de-absolutization through fact also furnishes the means of access to an absolute being; then we will have gained access to a truth that is invulnerable to correlationist scepticism.


In other words, correlationism is a false position inasmuch as it claims to be entirely agnostic about absolutes, because it lacks understanding of its own logical structure. You can only resist the idea that the correlation is an absolute on the basis of a principle that goes beyond correlationism. Speculative materialism is a correlationism that has such an understanding: to this extent, correlationism is source both of the problem and of the solution.

Meillassoux has begun to work through the consequences of his demonstration. But what edifice can you build on top of a principle so strange as the necessity of contingency? Firstly, this “principle of unreason” allows us to claim that something exists, i.e., to answer the question of why there is something rather than nothing. If there is nothing, there is no contingency; a pure void cannot be other than itself. Therefore, there must be something in order for there to be contingency (71). Secondly, the principle of unreason brings Meillassoux to argue that we can be assured of the universal validity of the principle of non-contradiction. A contradictory entity is already other than itself, leaving no space for its contingency (71). Finally, Meillassoux claims that we can speak of what exists in terms of “hyper-chaos,” a term that captures the raw existence of some thing purely contingent and non-contradictory (64). We thus gain, albeit in extremely limited terms, the ability to talk of a Real outside of the correlation. Furthermore, and this may be one of the most interesting claims, we can posit that facticity has always been the very source of the intuition of an outside in the first place: “The very idea of the difference between the in-itself and the for-us would never have arisen within you, had you not experienced what is perhaps human thought’s most remarkable power—its capacity to access the possibility of its own non-being, and thus to know itself to be mortal” (59). [End Page 171]

The Facticity of Research

Meillassoux’s argument merits considerably more scrutiny. In concentrating on the relationship to science, I must set aside other outstanding questions. We cannot probe for example the unresolved question of whether there are other possible defenses against idealism, one of the major chinks in the current state of his armor.

Meillassoux anticipates that he will eventually be able to make an argument that makes the transition between his concept of “hyper-chaos” and mathematics, but with the caveat that “our goal here was not to tackle this resolution as such. Our only aim has been to try to convince the reader not only that it is possible to rediscover thought’s absolutizing scope, but that it is urgent that we do so…” (128). He leaves us hanging. He has shown us that the very strength of the correlationist position is premised on actual access to an absolute, albeit a very minimalist, kernel, and thus that it shelters the hidden promise of a door beyond the perspectivism of language, culture, history and consciousness. What we can do, at this stage, is make a distinction between two manifestations of the human-world correlation. Firstly, we have the correlationism that is the source of the problem, Meillassoux’s point of departure. This extrinsic framing anchors everything relative to the human-world relation where it originates. However, speculative materialism also works through such a relation. It is precisely in some such relations that we can discover facticity, and are thus led along a new route to the absolute. We must be wary of imposing correlational forms upon scientific research, but at the same time, as a thoroughly modern enterprise, we must expect to find correlational forms endogenously operative within science, correlations that may through their facticity open a door beyond themselves. In short, while we gain purchase on speculative materialism through a radicalization of our appreciation of the correlation itself, this does not mean that correlationism itself, as an agnostic and relativist stance, is vindicated. If, as Meillassoux claims, scientific research can exceed correlationism, it is because its facticity shines through the particular kinds of human-world relation in which it is embedded.

I already argued that the first point would have to be to discard any assumption that the essence of science can be sought in the statements and texts that it generates. While these are important in many ways, the scientific endeavor is bigger than this. For climate science, we need to appreciate the role of simulations, specifically, of simulations that are too complex to be exhaustively described in the literature, of simulations that are operational, that are put to work, constantly modified and updated, the production and maintenance of which is just as much an output of research as are publications and documents, a global apparatus significant in its own right (Edwards 2010). These, along with the bodily dimensions of tacit knowledge, embody the inherently implicit, the fact that “we can know more than we can tell” (Polanyi 1983, 4; see also Collins 2010). The question of doing science justice [End Page 172] is not a question of the interpretation of statements, but of establishing the right kind of relationship with the multi-dimensional systems of investigation we know as research.

While any such practice-based account of science owes a debt to the tradition of STS (for example, Latour 1988, Knorr-Cetina 1999), it also strikes a chord with trends in contemporary Anglo-American philosophy of science. Sixty or seventy years ago it was more or less conventional for philosophers of science to foreground scientific theory as their core concern and to treat such theories as linguistic entities. But these days it is rare to foreground theory in these ways, except in a much more abstract sense of semantic structures (for example, Suppes 1960). In a parallel shift, philosophy saw moves to foreground the importance of experiment (Hacking 1983), and to undermine any sense in which theoretical laws and principles could be taken to have a self-sufficient coherence outside of the processes by which they get brought down to earth and articulated with concrete situations (Cartwright 1983). With a wave of interest in processes of modeling, many philosophers view science in terms of its actual practice, asking how models end up getting made, validated and manipulated in research projects (for example, Morgan and Morrison 1999; Teller 2001).

A practice-based view of science could capture the kind of thing that we would want to talk about when we talk about the correlation of research. It may not be consciousness-based or language-based, but it is still correlational. It still posits a domain of human-world relation which is the ineradicable site of research. Our move, to recap, must not be to dismiss this correlation as an illusion, but rather to radicalize it from the inside.

Following Meillassoux, we can talk of the facticity of these sites of research, the lack of reason why things are thus and not otherwise. It is important to clarify what this means. By denying the principle of sufficient reason, speculative materialism may appear to deny a basic tenet of coherent science. But there is no reason to think this would make any difference to how actual scientific research gets done. There is no need for scientists to be committed to the principle of reason as anything more than a methodological principle, one tool among many, which can be put to work with no assumption of metaphysical commitment.

The facticity of research captures the unreason behind things, manifest in the ever provisional and open nature of what gets carried out in the laboratory. I would suggest that facticity can be taken as a formulation of the most profound source of motivation inspiring the continuation of scientific work. While theoretical and practical principles can be drawn upon to support every part of the research process, these principles are themselves open and provisional. While there are many provisional reasons available, in the lack of ultimate reason behind how things are disposed in our investigations of them we find the intuition of reality itself. To recap, everything depends on the difference between taking this possibility of being otherwise as a lack of access to a hidden reason, and taking it to embody a truth that there is no such [End Page 173] domain. Indeed, ideas of such a hidden domain waiting to be “discovered” may turn out to be artifacts of a secularized theology of revelation, about which caution would be well advised (Rheinberger 2005, 409).

While Meillassoux gives a unique account of this facticity, it would be surprising if something similar had escaped the attention of others. Ian Hacking’s enigmatic “break” in his Representing and Intervening provides a quasi-mythological account of the concept of reality: “Once there is a practice of representing, a second-order concept follows in train. This is the concept of reality, a concept which has content only when there are first-order representations” (1983, 136). Hacking is not claiming that reality is just one representation among others. He is claiming that as soon as we have representing, which for him is always a material process, we have the possibility of a conflict between representations, manifesting the ever-present (potentially “open” and “factical”) possibility that things be otherwise. This tension, endemic to such relations, embodies a primary intuition of reality, a minimal but powerful inspiration for continued research. For both philosophers, who both take realism as a serious contemporary possibility, access to a concept of reality is nevertheless something that is gained through a correlational position, gained through the revelation that this relation harbors the possibility of being otherwise.

Similarly, Rheinberger, whose historical writings give us one of the most sophisticated accounts of scientific research in the making, points out that the key thing about scientific work is that it continually brings about new juxtapositions of systems of representation. Each new combination brings out the facticity of the correlation in new ways and continues to inspire with the real: “To bring alternative spaces of representation into existence is what scientific activity is about, and this is why the question of reality as an attribute of alternative representations, and the question of representation as an attribute of its alternative uses, will continue to stay at the center stage of the scientific enterprise” (Rheinberger 1997, 113). Scientific practice is not a singular endeavor, but a bringing together of multiple possibilities, each inflecting the others. The building of simulations in climate science involves the continual articulation of the computational system with the alterity of its others: mathematical derivations, software engineering theory, geological and physical theory, empirical data, experimental research.

Meillassoux risks blindness to the ways in which facticity emerges in actual research because he prioritizes mathematics, and with it carries the historical burden of a presumed unity of science. On the contrary, theories of scientific disunity are empirically compelling, especially for inter-disciplinary or trans-disciplinary fields such as climate science, and even suggest that it is the encounter and articulation between disunified sites that is the source of scientific efficacy (Galison 1997, 781; Hacking 1983, 183).

“The process of modeling,” says Rheinberger, “is one of shuttling back and forth between different spaces of representation. Scientific objects come into existence by comparing, displacing, marginalizing, hybridizing, and [End Page 174] grafting different representations with, from, against, and upon each other” (1997, 104). Science gets done in many ways and the juxtaposition and reconfiguration of scientific objects is always possible. Without an underlying foundation of unreason manifest in such situations it is hard to understand the desire to pursue things further, to open up new domains and new systems of exploration. Meillassoux presupposes the success of science. Hence the problem of our relation to it can be his point of departure. But scientists do not explicitly formulate a speculative materialist position. We would need to seek something of that nature implicit within their practice. The affective and implicit dimension of research might be the right kind of place to look: facticity as the heart of the drive to pursue things further?

The first stage is to show that the kind of correlation that we need to address with the speculative materialist argument is multidimensional field of practice. The second stage is to show that this field of practice can never ground the kind of agnosticism of which Meillassoux accuses correlationism. Meillassoux gives us a philosophical argument which aims to demonstrate the absolute facticity of that site. He provides the tools with which to unpick two very close but profoundly different interpretations of facticity: whether it is a manifestation of our finite human shortcomings in our attempts to reach the reason behind things, or, on the contrary, a manifestation of there being no such hidden reason. With the latter, he shows some ways in which the minimal contours of an ontology can be derived. But here we can pick up the thread with empirical research. If Meillassoux is right, we can explore the role of facticity in sites of scientific research. How do scientists encounter unreason? How do they, in articulating heterogeneous methods and fields, actively provoke the factical manifestation of things being otherwise? And what intuitions and motivations become grounded in this movement?


The most concrete conclusion I am able to draw is that it is interesting to look towards speculative materialism for new perspectives on science. As we live under the complex existential threat of climate change, this is not a matter of searching for novelty for novelty’s sake alone. There may be something here that provides perspective on the difficult problem of the relationship between science and society, or between science and politics.

My stance towards the practical site of research already presupposes that the proper relationship to science, the one that might “do science justice,” cannot be found in a public sphere. Science is more than discourse. Meillassoux’s attempts to move beyond philosophies of language would align with this. In seeking access to a new kind of absolute, Meillassoux attempts to locate a principle that would disrupt correlationists’ relativisms and fideisms, to gain some new and more concrete ground for a critique. Whether these philosophical maneuvers will really put a stop to debate is another matter, [End Page 175] but the intention is admirable. In responding to the question of the climate it is important to address climate activists as much as deniers, for there is too much faith among the former that greater volumes of information put forward into the public sphere will eventually weigh the argument down on their side. This is naïve (see, for example, Kahan, Jenkins-Smith, and Braman 2010). A better move might be to shift emphasis onto other grounds.

Facticity as the motivating unreason of the correlation speaks of the humility of science without reducing this to uncertainty estimates or probabilities. But doing science justice cannot be a matter of interpreting its statements literally. Doing so would place strange expectations upon it, and pander to the terms of the debate. Science is always debatable, but with Meillassoux’s analysis we might be able to discern between two types of debatability. On the one hand is the possibility of continuing to talk, of continuing to bring new arguments and evidence to the table, a recipe for inaction. On the other hand, we have the open possibility of being otherwise that grounds an intuition of reality, the inspiration of further scientific work, and the setting forth toward unprecedented events (Rheinberger 1997, 133–135). If it is recognized as absolute, as Meillassoux notes, this latter possibility takes us beyond the potential falsification of theories, and into the actual facticity of things (2009, 84). For science, it becomes a practical matter of bringing about new articulations, not a matter of continuing to return to the same positions of debate.

Being very hypothetical, any such argument runs the risk of becoming assimilated into a background of “yet more talk.” All I have been able to show is that perhaps these new trends in continental philosophy would offer something enlightening, that they may well open interesting doors, new possibilities in an area which is weighed down by the inability to produce conclusive discourse or decisive action. One potential starting point for opening these doors will be to look longer and harder at the accomplishment of science, with a self-critical view to the many manners and forms in which we might do science justice. And the ongoing development of Meillassoux’s thought may well prove an excellent platform for any such endeavor. [End Page 176]

Matt Spencer
Goldsmiths, University of London
Matt Spencer

Matt Spencer is an Associate Researcher at the Institute for Science, Innovation and Society at the University of Oxford. His research concerns the social dimensions of simulation use in the sciences. He is currently based in India in order to study the use of weather and climate forecasting in the water and disaster management sectors.


Badiou, Alain. Being and Event. Trans. Oliver Feltham. London: Continuum, 2007.
Bryant, Levi, Nick Srnicek, and Graham Harman, eds. The Speculative Turn: Continental Materialism and Realism. Prahran:, 2010.
Cartwright, Nancy. How the Laws of Physics Lie. Oxford: Oxford UP, 1983. 1st ed.
Collins, Harry. Tacit and Explicit Knowledge. Chicago: U of Chicago P, 2010.
Daston, Lorraine, and Peter Galison. Objectivity. New York: Zone Books, 2007.
Edwards, Paul N. A Vast Machine: Computer Models, Climate Data, and the Politics of Global Warming. Cambridge, MA: MIT P, 2010.
Galison, Peter. Image and Logic: Material Culture of Microphysics. Chicago: U of Chicago P, 1997.
Giere, Ronald N. Scientific Perspectivism. Chicago: U of Chicago P, 2006.
Hacking, Ian. Representing and Intervening: Introductory Topics in the Philosophy of Natural Science. Cambridge: Cambridge UP, 1983.
Harman, Graham. Quentin Meillassoux: Philosophy in the Making. Edinburgh: Edinburgh UP, 2011.
Heidegger, Martin. Being and Time. Trans. John Macquarrie and Edward Robinson. Oxford: Blackwell, 1978.
Kahan, Dan M., Hank Jenkins-Smith, and Donald Braman. “Cultural Cognition of Scientific Consensus.” Journal of Risk Research 14.2 (2010): 147-74.
Knorr-Cetina, K. Epistemic Cultures: How the Sciences Make Knowledge. Cambridge, MA: Harvard UP, 1999.
Latour, Bruno. Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard UP, 1988.
McWhorter, Ladelle, and Gail Stenstad, eds. Heidegger and the Earth: Essays in Environmental Philosophy. Toronto: U of Toronto P, 2009. 2nd ed.
Meillassoux, Quentin. After Finitude: An Essay on the Necessity of Contingency. Trans. Ray Brassier. London: Continuum, 2009.
Morgan, Mary, and Margaret Morrison. Models as Mediators: Perspectives on Natural and Social Sciences. Cambridge: Cambridge UP, 1999.
Pickering, Andrew. “From Science as Knowledge to Science as Practice.” Science as Practice and Culture. Ed. Pickering. Chicago: U of Chicago P, 1992. 1-26.
Polanyi, Michael. The Tacit Dimension. Gloucester: Peter Smith, 1983.
Rheinberger, Hans-Jörg. “A Reply to David Bloor: ‘Toward a Sociology of Epistemic Things.’” Perspectives on Science 13.3 (2005): 406–10.
___. Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube. Stanford: Stanford UP, 1997.
Suppes, Patrick. “A Comparison of the Meaning and Uses of Models in Mathematics and the Empirical Sciences.” Synthese 12.2/3 (1960): 287–301.
Teller, Paul. “Twilight of the Perfect Model Model.” Erkenntnis 55.3 (2001): 393–415. [End Page 177]

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.