This will be our reading for our February 24th meeting. You can find the details on our meetup.com page.
Meillassoux wants to describe two regimes of fiction and outline their philosophical importance. These two kinds of fiction both relate to the experimental sciences: science fiction and extro-science fiction, or SF and XSF. It will initially appear that extro-science fiction is merely a subcategory of science fiction, but he wants to argue there is a genuine distinction, that it is a “genre within a genre.”
1. Science Fiction and Extro-Science Fiction
1. The aspect of science fiction he wants to focus on is the imagining a future of science in which our knowledge and mastery of the real has greatly expanded. This expanded scientific capacity opens up unheard-of possibilities for man, but it remains recognizably scientific: “Every science fiction implicitly maintains the following axiom: in the anticipated future it will still be possible to subject the world to a scientific knowledge. Science will be transfigured by its new power, but it will always exist” (5).
2. In contrast, there is a “fiction of worlds outside-science.” He is not talking about worlds devoid of science, for example, fantasy worlds in which humans have yet to develop a scientific relation to the real. Rather: “By extro-science worlds we mean worlds where, in principle, experimental science is impossible and not unknown in fact” (5). (We could talk about the difference between fantasy worlds and science fiction worlds—despite the existence of magic in a fantasy story, physics still functions in a recognizable way and magic use still basically follows cause and effect rules. Harry Potter might be able to violate physics and create water out of nothing, but he has to cause the water to appear by casting a spell; otherwise, water acts like water just as it does in our world.)
3. Extro-science fiction imagines worlds in which science cannot deploy its theories or objects within them The goal of this essay is to give content to that simply negative definition of extro-science fiction, and it will do this by explaining what a world that it is in principle inaccessible to scientific knowledge would look like.
4. He is interested in extro-science fiction because of its relation to the problem of induction. Specifically, he is interested in the problem of the necessity of the laws of nature as outlined by Hume in his Treatise of Human Nature. For Hume’s empiricism, nothing guarantees the stability of the laws of nature. They are not logically necessary and past observations cannot tell us, with certainty, about the future.
5. Probably the most famous response to Hume was given by Karl Popper, but Meillassoux thinks Popper profoundly misunderstood Hume. Popper’s solution actually confused a problem of XSF with a problem of SF—Popper addressed a different problem than Hume did. Hume raised a problem of XSF, while Popper resolved an SF problem. Kant, in contrast, did not misunderstand Hume and did respond to him on the proper terrain, by “fictioning” a world in which science becomes impossible. However, Meillassoux thinks Kant’s answer still fails because it had an insufficiently developed XSF imaginary. A “more acute sense” of XSF gives us a third response to Hume’s problem.
2. Two Billiard Games: Hume and Asimov
a) Formulation of the Problem
6. In Hume’s discussion of causal necessity, he describes a billiards game in which the laws of collision no longer hold:
“When I see, for instance, a billiard-ball moving in a straight line towards another; even suppose motion in the second ball should not by accident be suggested to me, as a result of their contact or impulse; may I not conceive, that a hundred different events might as well follow from that cause? May not both these balls remain at absolute rest? May not the first ball return in a straight line, or leap off from the second in any line or direction? All these suppositions are consistent and conceivable. Why then should we give the preference to one, which is no more consistent or conceivable than the rest? All our reasonings a priori will never be able to show us any foundation for this preference.”
7. In this imaginary scene, Hume asks what guarantees for us, and also what convinces us, that physical laws will still be valid in the next moment, since neither experience nor logic can give this assurance. There is no logical contradiction in imagining a future modification of laws, and no past experience can eliminate this possibility.
8. One could respond that science allows us to predict future phenomena, like eclipses. These predictions, however, rest on the hypothesis that current laws will persist, and nothing but “good sense” assures us this is the case. Hume’s answer to this problem is that habit convinces us of future stability, but this answer did not satisfy those who came after, especially Kant and Popper.
9. Meillassoux begins with the most recent solution, from Popper’s The Logic of Scientific Discovery. The solution appears quite simple. If we ask Popper what guarantees the behaviour of the Humean billiard ball, he says nothing can guarantee it, but that this is a good thing and there is nothing fantastical about it. Our predictions about the future are falsifiable hypotheses. This falsifiability is what explains the dynamism of science, and why we cannot guarantee the future, because our hypothetical predictions can always be falsified.
10. In Objective Knowledge, Popper gives three examples of “established laws” that were refuted. For example, the “law” that every living being perishes was refuted “by the discovery that bacteria are not bound to die, since multiplication by fission is not death,” or the “law” that bread nourishes was refuted by ergotism (which results from eating rye bread infected with a fungus).
11. In the billiard ball example, the balls could adopt new behaviours for a variety of reasons: we could magnetize them, for example. Popper’s response to Hume is that every event is compatible with either a current or future state of science. This does not answer the problem as formulated by Hume. First, this solution moves entirely within a science fiction imagination. The physics of the future may be even more unexpected and unrecognizable than relativity was to Newtonian physics, but it will still be physics.
12. Popper says our theories may be falsified in the future: his problem is an epistemological one. However, Hume’s problem is an ontological one. It is not about the stability of theories but the stability of processes. In Popper’s framework, chemically identical loaves of uninfected bread will never induce ergotism. If such a change were to happen, it would not be a revolution in theories adapting to new experiments, but as the result of the collapse of physical laws. Meillassoux says, “If you abolish every constancy in the results of an identical experiment, the principle of experimentation—the reproduction at will of the phenomenon under the same conditions—will collapse and with it the possibility of the natural sciences, whether their theories are deterministic or probabilistic” (16-17).
b) Professor Priss’ Crime
13. The hypothesis of a future in which science is impossible is Hume’s real problem. Popper’s problem is a problem of science fiction; Hume’s problem is one of extro-science fiction, the fiction of a world which is too chaotic to allow for scientific theory. This is what hints at the philosophical stakes of SF and XSF: it confused Popper and allowed him to think he resolved Hume’s problem.
14. Isaac Asimov’s story “The Billiard Ball” illustrates the difference here quite well. The story is about a murder carried out by the unforeseen trajectory of a billiard ball. Importantly, this story can only be understood in Popper’s science fiction framework, and not in Hume’s XSF framework.
15. The story has two characters: a theoretical physicist named Priss and an engineer named Bloom. Priss’ theories have earned him acclaim in academia, but Bloom’s inventions have been him rich and famous in the wider world. Priss is violently jealous of this. Priss develops a theory of anti-gravity, but says it cannot be practically applied because it would require infinite energy. Bloom sets his mind to proving this claim of impossibility wrong. At a press conference, Bloom invites Priss to active his new anti-gravity machine, which will cause a billiard ball to levitate. Priss does so, and Bloom is killed by the ball’s unexpected path.
16. In an XSF world, there would be nothing more to say about this aberrant event. However, this is a science fiction story, and eventually, we discover the scientific cause of the disaster. The event was in fact unforeseen, but it was not in principle unforeseeable, and the physicist Priss offers a proper scientific explanation of the ball’s path.
17. There is an apparent general conclusion to be drawn here: only science fiction allows the construction of a coherent if fanciful storyline. Stories can be told because they deal with ordered totalities, or what Meillassoux calls worlds, even if they are governed by a different order. Individuals are capable of acting and planning within them. In an XSF world, however, it seems there can be no order and therefore no story told. To the extent that is true, we cannot speak of extro-science fiction worlds, because a world which could not give a place to science would not be a world but a pure chaos.
18. This is Kant’s response to Hume: if laws were not necessary, there could be no consciousness. There would only be a pure manifold with no cohesion. Meillassoux intends to challenge this thesis, and show that an XSF world is possible.
3. The Transcendental Deduction and the Three Types of XSF Worlds
a) Kantian Rejection of the Fantastical Billiard
19. In The Critique of Pure Reason, Kant’s response to Hume appears in the “objective deduction of the categories.” The understanding, responding to a sort of logical necessary, produces the categories that individuate and outline objects for us. The categories include contingency and necessity, unity and plurality, and cause and effect.
20. The goal of deducing the categories was to legitimate their application to experience. This legitimation requires some work, because these categories are “universal forms”, while experience deals with particulars. Deducing the category of causality in particular means resolving Hume’s problem: under the same circumstances, identified causes will always produce the same results. Kant wants to legitimate our belief in the necessity of physical laws, but without relying on a baroque metaphysical vision like Leibniz’s.
21. Kant said we could never perceive Hume’s erratic billiard ball because the contingency of natural laws would make object-consciousness impossible. If we could imagine Hume’s ball, then it would simply be the foreground of a background which remains stable. If the laws failed in the case of the balls, they would also have to fail for the world in general, rendering impossible any subjective representation of the world. In other words, Kant’s response to Hume is that Hume illegitimately dissociates the conditions of science from the conditions of consciousness. For Kant, reason cannot survive the absence of consciousness:
“This proves the impossibility that such a collapse of science and of natural laws may become manifest to us one day; we will never see the Humean ‘billiard scene,’ not because it is absolutely impossible that our world would collapse one day—only a speculative metaphysician can affirm this impossibility in an absolute way—but because the collapse of this world would be ipso facto the collapse of every world-form as well as that of the consciousness capable of witnessing this spectacle.” (27)
22. Meillassoux’s goal here is to restore not the letter but the spirit of Kant’s argument, which involves three steps. First, suppose that laws stop governing the given. Science would become impossible, but we could never perceive this, except perhaps as a dream. The difference between perception and dreaming, for Kant, is directly related to constancy. It is the difference between objective representations (arising from experience) and chimerical representations (arising only from imagination), or the difference between representations that are ordered by the categories (here, especially causality) and those which are not.
23. Second, consider that if the laws disappeared, the real would not even have the consistency of a dream. A real lawless world would be too unstable to even allow time for the differentiation of the random changes, since every stability would be absolutely ephemeral.
24. Third, since even temporal continuity has been broken, self-consciousness itself could not survive to witness this. Reality would become even less real than a dream.
25. Kant’s argument is a factual demonstration: since the contingency of laws would involve the abolition of representation and self-consciousness, the very fact that there is representation refutes the Humean hypothesis. Importantly, this also seems to refute the very possibility of extro-science fiction as a literary genre: “Such an imaginary seems bound to be reduced to the monotony of a pure disorder at the heart of which nothing subsists and nothing is distinguished from nothing” (31).
b) The Possibility of Non-Kantian Worlds
26. However, it is that claim that suggests a weakness in the Kantian solution: there is nothing that actually prevents us from imagining extro-science worlds that are more stable than those described by Kant. These would be worlds not subject to necessary laws, but still stable on the whole. Or in other words, “Why should a lawless world be, without fail, frenetically inconstant?” To this, Meillassoux replies that a world obeying no law would have no more reason to be chaotic than it would to be ordered. Nothing could prevent the composition of a global order containing small details that could “run out of control.” In other words, the deficiency of the transcendental deduction is that it does not push hard enough on its XSF imaginary.
27. Further, Kant’s claim that science and consciousness have the same conditions of possibility—the necessity of laws—is itself faulty, because we can imagine three types of worlds that contradict it.
28. Type 1 worlds are worlds that are irregular, but this irregularity does not affect science or consciousness. These are not XSF worlds in the strict sense, because science still functions in them, but they would still contradict the thesis that the strict necessity of laws is a condition for the possibility of science. In this world, causeless events would be too rare to be of any significance.
29. Science would be unaffected because these events could only give rise to testimony, and would be irrelevant to any protocol of observation. The irregular events would always be things that happened to a friend-of-a-friend. Consciousness would also be safe, because the event could be distinguished from a dream or a hallucination in a few ways. Dreams involve sleep and waking up, and hallucinations can be tied to specific pathologies. The criterion of intersubjectivity would also eliminate the possibility of hallucination.
30. Since type 1 worlds are thinkable, this proves that neither science nor consciousness require the strict universality of causality.
31. In a type 2 world, irregularity abolishes science, but not consciousness. This is a genuine extro-science world. In these worlds, no sphere would be safe from a-causal disorder. Identical laboratory experiments would be capable of producing diverse results, ending the possibility of a science of nature. However, in this world, daily life could continue, because there would still be a relative stability. Strange events would be common enough to prevent general theories, but not daily action.
32. It would be a world in which we could only chronicle things: “[F]rom this date to that date, the nature ‘of the laboratory’ ceased to be relativist and regressed toward Newtonian dynamics,” or “from this date to that day, a genuine ‘renewal’ of quantum physics took place, particularly in the laboratories of the southern hemisphere” (36-37).
33. We need to remember that no manifest irregularity could prove that there is no law underlying this apparent disorder, but the person seeking this order would be just as eccentric as those seeking a quantitative law capable of explaining human history in our world.
34. It might be possible to find a statistical regularity for these events, analogous to the statistic regularity of humans breaking traffic laws. This would seem to be the beginning of a new science of nature. We need to examine this analogy between social and natural regularities by adding a historical element. Consider a person in 19th century France measuring the probability of carriage accidents in Paris. Now imagine that he knows that in the 20th century, the probability of carriage accidents would tend towards zero: he might assume that carriage safety had been greatly improved, because he would be unaware of the car. Meillassoux says, “Thus social regularity, which allows us in the short and medium term to build on the quantifiable probability of the behavior of others despite its individual unpredictability, is coupled with the possibility of a historical change on the largest scale: an unpredictable change in a more profound sense, because it is impossible to subject it to any quantifiable law” (39).
35. But such a massive change does not eliminate all social regularity. In the case of science, a type-2 world would have a similar issue: the possibility of changes so large that they escape probability, but would still allow consciousness to function.
36. Finally, a type-3 world, in which neither science nor consciousness would be possible, as in Kant’s objective deduction.
37. Of the three types of extro-science worlds, two of them contradict the transcendental deduction, and one constitutions a genuinely extro-science fiction world, which he calls the XSF-2 world.
38. The point of this is to ask if XSF can be a narrative genre that could compete with SF.
4. Extro-Science Fiction and Narration
39. This section is about XSF novels.
a) Three Procedures
40. . The problem with XSF novels, which at first sight condemns them to being quite rare, is that they start with what usually has to be excluded from narration: not merely pure arbitrariness, but arbitrariness that could appear at any moment. In SF novels, the reader grants the author quite a few fanciful postulates, but they have to be adhered to. If the rules changed without reason, then the story would become boring and nonsensical. How could a story, in which events took place for no reason, be structured?
41. An XSF story has to obey two requirements. First, events must take place that no real or “imagined” logic can explain. Second, science has to be an issue in the story, though in a negative way. It is a world in which science (in one or all disciplines) becomes impossible. Or, even more radically, a world in which the absence of the possibility of scientific explanations is intensely felt. These are the traits which distinguish XSF stories from fantasy or Lewis Carroll-style nonsense. In these stories, the absence of science is not felt because it is replaced by a different logic.
42. XSF does not have “a coherence of change,” but must still compose a story. There are three types of solutions to this problem. The first solution involves introducing a single break, a unique catastrophe. Robert Charles Wilson’s Darwinia is an example of this. In 1912, Europe disappears, leaving behind a radically different continent. While this event initially defies all scientific explanation, eventually one is found: the Earth this event took place on is not the real Earth, but an archived version of the Earth. A hitherto unknown powerful force caused the event.
43. The second solution is to have multiple breaks in order to verge on nonsense. These worlds tend to be comedic, and the example here is The Hitchhiker’s Guide to the Galaxy with its infinite improbability engine. However, in this case, probability still functions, and an underlying coherence remains. Importantly, the machine can be started and stopped at will, and so the crazy events are not truly a-causal.
44. The third solution would be a novel that takes place within an uncertain reality, a world that becomes increasingly unfamiliar. Like the second solution, there would be multiple breaks, but the effect is of progressive disintegration rather than comedy. The example here is Philip K. Dick’s Ubik, a world suffering increasing incoherence. Phonebooks randomly become out of date, and a recently assassinated man’s image appears in random places. This creates a nightmarish world, the flip side of Hitchhiker’s comedy. Eventually, we discover we are dealing with the psychic world of individuals who are in cryogenic suspension.
45. The problem with all three candidate solutions is that in the end, SF explanations are reintroduced.
b) An XSF prototype
46. Meillassoux has found at least one example of a genuine XSF novel, which proves that the genre can exist and even have popular success: Rene Barjavel’s Ravage (Ashes, Ashes in English). Like the previous examples, Ravage begins with an SF context, and adds a foreign logic. But unlike the previous examples, the events actually do escape an SF context of reasons.
47. The story portrays a world in which electric stops working, or is at least is no longer manifest. Barjavel never offers an explanation for this event, though the characters search for both scientific and theological explanation such as sunspots or divine punishment. Nothing ever confirms the character’s conjectures. The two main explanatory speeches given by characters amount to admissions of ignorance. The first is given by a scientist, lamenting the fact that all theories have been overthrown. The second is given by an avatar of “good sense” who says “we know nothing.” Nothing is excluded as a possible explanation, and nothing is confirmed.
48. As already stated, an XSF-2 world cannot completely rule out the presence of an underlying law; “The key is that the very idea of explanation is deprived of its stakes and that the inhabitants of this world have all their time taken up by the vagaries of the environment that has become unpredictable and unrecognizable” (52-53).
49. Ravage was written in the 1940s, in a context of anti-modern “return to the countryside” type ideology. We should not ignore this inglorious context, but remember that a successful work always surpasses the context it was produced in. Part of what makes Ravage interesting is that the disappearance of electricity is never explained or even interpreted in terms of Barjavel’s pastoral fantasies.
50. Even more importantly, the novel “transposes onto itself” the historical catastrophe of 1940 and the troubles with followed, such as the extinguishing of lights and the blackout imposed on Paris after 4:00. The point to take from this:
“This intersects with the comparison I tried to sketch out between type-2 worlds and the radicality of historical contingencies: the soft ground of the vanquished nation is transformed into the soft ground of changing nature. The political stupidity of the storyline matters little then; it cannot eliminate the originality of the tale: that it is an authentic example of XSF, a controlled tale in a world without substance.” (56)
51. Hence, it seems possible that XSF can exist as its own genre.
52. Meillassoux’s final point is that XSF can go beyond being a simple adventure tale:
“[S]tarting from traditional science fiction, we can decompose it by tilting the world toward extro-science and pursuing this enterprise of degradation toward a less and less inhabitable world, making the tale itself progressively impossible, until we isolate certain lives that are tightened around their own flow in the midst of gaps. Life mentally experience itself without science and, in this ever more accentuated divergence, perhaps discovers something unprecedented about itself or about science. An eidetic variation pushed to the point of suffocation, self-experience in a non-experienceable world. A precarious intensity would plunge infinitely into its pure solitude, with only an environment of rubble in which to explore the truth of a worldless existence.” (57)
53. Here is my best attempt at interpreting this paragraph. The “eidetic variation” he mentions is a play on Edmund Husserl’s “eidetic reduction,” his method of stripping away the roles that memory, expectation, and existing knowledge play into our experience of the world so as to focus on the moment of experience itself. XSF allows us to imagine a world of chaos in which individual lives are still led, a chaos which reveals something about both experience and the conditions of science; they are not as tightly intertwined as Kant would have us think. The conditions of science can be eidetically reduced away from the conditions of consciousness. But likewise, the conditions of consciousness are not dependent on the conditions of science. All this leads me to ask: does mathematized science allow us to escape our linguistic “worlds,” as described by Cavell?