§4.1 — Explicitly acknowledged ‘network problems’ long pre-exist the electronic era, by centuries, if not millennia. They constitute a guiding thread within what has been called ‘the tradition of spontaneous order’. Order is spontaneous if it solves a coordination problem without appeal to any organizational element at a higher – or super-social – level. Spontaneous order, social self-organization, or immanent social process, is thus counter-posed to transcendent social design (from above, or beyond), and to its corresponding theoretical justifications, which amount to a social metaphysics, typically serving concrete functions as guiding ideologies of super-ordinate control. As long recognized by its opponents, the assertive recognition of this theoretical conundrum cannot be practically dissociated from an implicit political stance.
§4.11 — In the absence of superior guidance, solutions to coordination problems have to emerge, out of – and as – games. This is only to say that cooperation between the agents involved cannot be presupposed, but has to arise from their interaction, if it is – indeed – to appear at all. To assume altruism or solidarity in such cases is a failure to think the problem at all. Coordination is the explanandum. The collectivist dogma is not an answer, therefore, but an alternative to the question. An answer is a trust machine, for which Bitcoin is the model. It is strictly complementary to a minimal presupposition, that of trustlessness (for which it is the solution).
§4.12 — Byzantine generalization extends to the very limit of network communication problems, to the difficulties of establishing coordination within radically dispersed (and thus zero-trust) multiplicities, encompassing non-idealized societies / populations of every variety. It is not, of course, that the concrete existence of trust is simply denied, only that it is rigorously thematized, as a social product requiring explanation – and what counts as an ‘explanation’ cannot, whether overtly or covertly, merely presuppose what it is called upon to explain. (This demand, as we have seen, is already totally – even ultimately – controversial.) If there is trust, there has to be a trust engine, conceived without pre-existent bonds of trust as a part. At the most abstract level, therefore, this is a topic that would have been familiar to the thinkers of the Scottish Enlightenment, as to all those participating productively within the theoretical tradition of spontaneous order. It is exactly this same problem of decentralized coordination (in the absence of any transcendent contribution provided by assumed altruism or common purpose) that has been the essential guideline for realistic social analysis within the ‘whig’ lineage of descriptive liberalism, exemplified most famously by Adam Smith’s figure of the ‘invisible hand’.
§4.13 — Evolutionary biology, as a science of emergent networks, has engaged very similar problems from its inception, often with identical tools. This is especially evident in the modeling of ecological equilibrium within large, long-term biological system dynamics, in which the absorption of extinctions defines the ‘true network’ (by the absence of indispensable nodes). A more recent attempt to formalize such coordination is found in the game theory of John von Neumann, which has itself been effectively applied to biological networks at a variety of scales. The latest – and still rapidly self-transforming – incarnation of this tradition can be seen in the science of ‘complex adaptive systems’ as exemplified by the research programs of the Santa Fe Institute. In each case, the defining theoretical object is emergent coordination, in which no appeal to any centralized, superordinate, or orchestrating principle is made, unless this can be identified with the system itself, as such. The target of such researches is transcendental order, as captured by the immanent rules of distribution (which are ultimately indistinguishable from the real correlate of mathematics).
§4.14 — Games, strictly understood, therefore, arise under the minimalistic assumptions tolerable to an analytical anarchism, that is: after the methodical subtraction of all presumed coordination, or the conversion of such presuppositions into formal theoretical problems. The implicit critical impulse driving the construction of such research programs is evident. That which might have been asserted, as a transcendent principle, or metaphysical dogma, is to be instead explained, as an immanent, emergent, or ‘evolutionary’ outcome. Whether explicitly understood in such terms, or not, every such enterprise is a regional application of critical philosophy. Spontaneous order is the correlate of critique. The solution, however, cannot correspond to a philosophical thesis (of any traditional type), or even to a ‘machinic proposition’ – an engineering diagram, simulation, or protocol – but can only emerge as the synthetic product of such a proposition, when executed. The notion that coordination problems (of significant complexity) can be anticipated in detail by a process of pure ratiocination is a philosophical disease, of recognizably pre-critical type. The idea of the diagonal is not the diagonal.
§4.15 — If the discovery of spontaneous order as a problem corresponds to the execution of critique, it can be formalized through a diagonalization matrix. When the pre-critical opposition of centralized coordination to uncoordinated dispersion is tabulated, it draws a graph isomorphic with the Kantian schema. At the level of the abstract formalism, this latter is echoed with such extreme fidelity that we can borrow from it freely, while switching its terms through a simple hash. Substitute centralization for analysis, decentralization for synthesis, order (coordination) for the a priori, and disorder for the a posteriori. As with the Kantian original, the first diagonal term fails – there is no centralized disorder, any more than there are analytic a posteriori judgments. Centralization is an intrinsically anomalous distribution, necessarily threatened by the prospect of a fall into disorder. Its complementary conception, ‘simple anarchy’, is no less invulnerable to theoretical dismissal. The previously acknowledged terms, centralized order, and decentralized disorder (like the analytic a priori, and synthetic a posteriori) are therefore preserved. Common sense is not abolished, at least, not initially. In the exact formal place of Kant’s invention/discovery of the synthetic a priori, the critique of coordination, too, generates a viable diagonal product – decentralized order. (Quod erat demonstrandum). This is the Great Oblique worked by all realistic social theory since the inception of the modern epoch.
§4.16 — In Cyberspace, the tradition of spontaneous order has been massively accelerated. Classical coordination problems have been reformulated as experimental fields, opened for exploration by the emergence of increasingly-powerful computer simulation tools, and consolidated as practical solutions through the implementation of cryptographic systems and P2P platforms. Philosophical reflection has, to a very considerable extent, been side-lined by technical applications reinforced by far superior criteria of evidence, which is to say: practical demonstration. In the age of electronic information technology networks can be tested as products, and it is possible to ask of a complex idealization, as never before, does it work? This transition, through computer simulation, from explanation to implementation, registers a process of technological envelopment without definite limits. It corresponds to an interminable tide of disintermediation that established institutions of intellectual authority have not yet begun to fear enough.
§4.17 — While institutionalized philosophy has tended to lag the network trend, rather than pioneer it, something that might reasonably be called ‘abstract network theory’ has nevertheless arisen to become a guiding philosophical theme since the mid-20th century. Among those modes of philosophical writing that have been most influential upon the wider humanities and social sciences, this attention to the distinctive characteristics of networks has been especially notable. Appropriately enough, this ‘discourse’ – or dispersed conversation – has no uncontroversial center of authority, stable common terminology, shared principles, or consistent ideology. Its rhetoric tends to systematically confuse advocacy with criticism, and both with analysis, as it redistributes partial insight around a number of relatively tightly-interlocking social circuits (which are overwhelmingly dominated by a familiar set of theoretically-superfluous moral-political axioms). Yet, however deeply regrettable this concrete situation might be considered from the perspective of austere theory, it cannot be simply wished away. Intellectual production itself occurs within networks, and those with greatest promotional capability are among those least optimized for pure intellection.
§4.18 — The cultural systems in which the philosophical (and sub-philosophical) formalization of radically decentralized – or ‘true’ – networks has emerged, through a multi-decade self-reflexive process, are eccentrically articulated, very partially self-aware, and only weakly integrated. Yet even in these noisy and deeply compromised circles, cross-cut by vociferous extraneous agendas, and subjected only very weakly to a hard reality criterion, convergence upon the rigorous conception of a model network has been inexorable. An especially influential example is the rhizome of Gilles Deleuze and Félix Guattari, which provides philosophy with its most rigorously-systematized account of acentric and anorganic order.
§4.19 — A network, in this sense, has no indispensable nodes, entitling it to the adjective ‘robust’. Once again, the Internet – at least in its idealized conception – exemplifies such a system. It is typical, at this point, to recall the origins of this ‘network of networks’ in a military communications project (ARPANET), guided by the imperative of ‘survivability’ realized through radical decentralization. As will be repeatedly noted, on the crypto-current the security criterion is primary, providing system dispersion with its principle. This suggests, as an ideological generalization, that there is no basic ‘trade-off’ between liberty and security. Rather, it is through security (alone) that liberty establishes its reality. The term “crypto-anarchy” condenses this conclusion into a program.
§4.191 — Such invocations
of the strategic investment in distribution and redundancy, however predictable,
remain conceptually productive. They are especially valuable as a corrective to
modes of discourse – typical among contemporary humanistic studies – which tend
to haze the harsh selective or eliminative function of critique into a vapid
metaphor. It is the military ancestry of the Internet that is tacitly
referenced in the celebrated maxim of John Gilmore (1993): “The Net interprets
censorship as damage and routes around it.” In order to save command-control,
it was necessary to fundamentally subvert it.
 Barry Norman’s ‘The Tradition of Spontaneous Order’ (2002) lays out the intellectual history of the idea, within a primarily legal and economic context: http://www.econlib.org/library/Essays/LtrLbrty/bryTSO.html
Whe abstracted beyond its socio-historical exemplification, and apprehended as a cosmic-philosophical principle, spontaneous order refers to immanent coordination within dispersed collectivities, making it approximately synonymous with ‘the science of multiplicities’ in general. It designates the systematicity of the system, which can be alternatively formulated as an irreducibility without transcendence, or an immanent non-locality. Whether philosophical, or colloquial, the association of spontaneity with notions of vitality is ultimately arbitrary. Spontaneity is not life, unless ‘life’ is conceived as the irreducible operation of the multiple as such.
 Public Choice Theory serves as a reliable proxy for minimalistic, game-theoretical axioms in their directly political application. In her book All You Can Eat: Greed, Lust and the New Capitalism (2001), Linda McQuaig cites a short fable by Amartya Sen, designed to satirize the Public Choice approach: “‘Can you direct me to the railway station?’ asks the stranger. ‘Certainly,’ says the local, pointing in the opposite direction, towards the post office, ‘and would you post this letter for me on your way?’ ‘Certainly,’ says the stranger, resolving to open it to see if it contains anything worth stealing.” Notably, in contenting itself with a satirization of disciplined moral skepticism, this line of criticism is brazenly unembarrassed about its dogmatic structure. Its rhetorical brilliance diverts from its substantive socio-historical meaning, which is the acknowledgment that solidarity is premised on the absence of a question. In other words, it demands, finally, the docile acceptance of an inarticulate presupposition. In this demand is to be found its irreparable weakness, shared by all cultural commitments – typically religious in nature – that are grievously wounded by the very fact of coming into question. A comparable defense of transcendent altruism founded upon the pre-delimitation of critique will be seen in David Graeber’s ‘everyday communism’. There, too, fatal injury coincides with the beginning of skeptical thought. Since, in this framework of collectivist moral norms, the investigation is the real crime, an intrinsic inclination to totalitarianism proves difficult to control.
There is one, and only one, coherent rejoinder available to the Left on this point. The table has to be reversed, in order for autonomous individuated agency to occupy the role of the explanandum, with its production and reproduction within the collectivity determined as the critical problem. Naturally, the production of individuation within complex adaptive systems would not satisfy this demand, since the presupposition of any such analysis is uncoordinated multiplicity. The Left is compelled to maintain that war is not ‘the father of all’. It is darkly amusing, then, that we continue to argue about it.
 In Smith’s most widely-cited words, taken from the Wealth of Nations, the repudiation of assumed altruism is explicit: “It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own self-interest. We address ourselves not to their humanity but to their self-love, and never talk to them of our own necessities, but of their advantages.” It is only through this theoretical self-restraint that the emergence of social cooperation is critically explained, rather than dogmatically asserted. It is important to note, however, that the recognition of the problem is not in itself a solution. The solution is not a philosophical reservation, but a machine. In Smith’s case, of course, it is the market mechanism, understood as a catallactic generator, or engine of spontaneous order, that represents the general (abstract) form of social solutions to novel coordination problems, in a way that remains mainstream within economics up to the present day. When grasped within this framework, it is at least tempting – if perhaps not strictly compelling – to suggest that any purported successor to the market, capable of usurping the role of the market, could only itself be a new iteration of the market. The most prevalent contemporary radical (left) alternative requires that coordination problems, rather than being solved, are instead terminally dispelled. Among those unfashionable vestiges of the Marxist tradition which emphasize the inheritance of capitalistic management practices by post-capitalistic social regimes (as exemplified by Lenin’s Taylorism), there is at least the promise of collective action solutions taking the form of legacy cultural resources, although without solid prospect of further innovation in a post-competitive milieu.
 The formal application of game theory to evolution dates back to the early 1970s, pioneered by John Maynard Smith and George R. Price, through the theorization of evolutionary strategies (see ‘The Logic of Animal Conflicts’, 1973, http://adsabs.harvard.edu/abs/1973Natur.246…15S). Since the notion of ‘strategy’ has an ineliminable telic inflection, this mode of evolutionary formalization can be philosophically-conceived as a diagonalization of biology. A video interview with Smith on the topic can be found here: http://www.webofstories.com/play/john.maynard.smith/42;jsessionid=7E3CA894713888028A8D7066FB3F263E
The preparatory theoretical work of W. D. Hamilton is especially notable in this regard, since it is explicitly focused on the application of Darwinian intellectual tools to subtract the metaphysical pre-supposition of altruism. The rhetorical provocation of selfishness, as seen most prominently in the Hamilton-based neo-Darwinian synthesis of Richard Dawkins, testifies to this project. The ‘selfish gene’ poses altruism as a problem, which Hamilton had already productively addressed. This is how transcendental critique reaches social biology, whose compressed name – socio-biology – was, from its inception, a political scandal. Collectivist metaphysics was not slow to recognize the profound intellectual menace – at once ancient and scientifically-renovated – that it faced.
 The Santa Fe Institute gathers cross-disciplinary researchers for the study of far-from-equilibrium systems and their emergent behavior, characterized by path dependency, sub-optimality, and massive informational imperfections. Of special relevance to the discussion here has been the pioneering of ‘Complexity Economics’, associated in particular with the work of Brian Arthur, which provides a disciplined critique of equilibrium models in neoclassical economics. Markets approach equilibrium only in the way that all working machines approach an entropy maximum, without this limit being ever – or even approximately – actualized. See the institute’s website at http://www.santafe.edu/
The concrete socio-historical application of such thinking found in Manuel De Landa’s (2000) A Thousand Years of Nonlinear History (see bibliography).
 Between ‘descriptive liberalism’ and ‘analytical anarchism’ – despite the apparent terminological escalation – there is no distinction of serious theoretical significance. ‘Order out of chaos’ (or at ‘the edge of chaos’) is the consistent explanatory horizon.
 The most influential proximal ancestor of current network theory (by far) has been connectionism. This intellectual movement, characterized by an extreme interdisciplinarity, introduced the problems of electronic engineering and distributed information systems into the study of biological, psychological, and social phenomena. Its name was coined by neural net-theorist Donald Hebb, in the 1940s. Rising to prominence from the early 1980s, connectionism has generalized the study of neural networks (or parallel distributed processing) in suggestive coincidence with the rise of the Internet. Its research orientation has been guided by the proposition that complex adaptive behavior is better explained by the pattern of connections between simple units than by complex properties inherent within concentrated entities of any sort. At a larger scale, connectionism updates the diffuse multi-century current of empirical associationism, accommodating it to the prospects of technological actualization within electronic networks. It is thus marked by a comparative disregard for elaborate symbolic structures, and for highly-organized internal states more generally. The atomic elements of connectionist analysis are linkage modules, supporting emergent systemic behavior. Like cybernetics before it, connectionism outgrew its own disciplinary contour, and dissolved into the sophisticated common sense of the world. (Generalized dominion can be difficult to distinguish from disappearance.) Subsequent attempts to specify a definite ‘theory of networks’ have been programmatically vaguer, resorting typically to some notional gestures of obeisance in the direction of mathematical graph theory. Much of this work has been seduced by literary temptations, bringing its enduring theoretical productivity into serious question. Most damagingly, in respect to its own capacity for conceptual genesis, the primary analytical discipline of connectionism – programmatic dismantling of mysterious essences into distributions – has been increasingly neglected, and replaced by an arcane cult of whole ‘objects’ withdrawn from the menace of disintegration.
 Axioms are independent, formally-articulated assumptions. Since each axiom is a basic presupposition which cannot (therefore) either be derived from, or support, any other, a set of axioms is an irreducible multiplicity. Consequently, an axiomatic has a characteristic mode of variation, based upon composition from wholly independent parts, which can be added or subtracted with comparative freedom. Because axioms are bedrock elements, their selection demands an irreducible experimentalism. (As a matter of logical necessity, the systems they compose can never determine the characteristics of a missing axiom.) Within a social context, the pursuit of minimal axiomatic systems is ideologically charged, since it corresponds to a contraction of public purposes. The historical disintensification of capitalism has proceeded, as Deleuze & Guattari note, by way of an axiomatic proliferation. Addition of axioms is the way capital has been socially compromised. From a Francophone perspective, this tendency appears as a resilient teleological structure. When the same prediction is extended to Anglophone cultures, in which the recession of classical liberalism remains far more seriously contested, less confident conclusions are advisable.
 Rigorous transcendental formulation of the model network – with all the conceptual ironies and traps this involves – finds its highwater-mark in Deleuze & Guattari’s essay ‘Rhizome’ (which introduces the second volume of Capitalism & Schizophrenia, A Thousand Plateaus). A ‘rhizome’ acquires facile (contrastive) definition through its distinction from the ‘arborescent’ schema of hierarchical sub-division. It proposes an absolute horizontality, approached through dimensional collapse, which opposes the rhizome to the tree, reticulation to hierarchical structure, and contagion to heredity. “In truth, it is not enough to say, ‘Long live the multiple’, difficult as it is to raise that cry. The multiple must be made, not by always adding a higher dimension, but rather in the simplest ways, by dint of sobriety, with the number of dimensions one already has available – always n – 1. (the only way the one belongs to the multiple: always subtracted). Subtract the unique from the multiplicity to be constituted; write at n – 1 dimensions. A system of this kind could be called a rhizome” (ATP 6). When analytically decomposed, the rhizome exhibits a number of consistent, distinctive features. “1 and 2. Principles of connection and heterogeneity: any point of a rhizome can be connected to anything other, and must be. … 3. Principle of multiplicity: it is only when the multiple is effectively treated as a substantive, ‘multiplicity’, that it ceases to have any relation to the One as subject or object, natural or spirtitual reality, image and world. Multiplicities are rhizomatic, and expose arborescent pseudomultiplicities for what they are. … 4. Principle of asignifying rupture: against the oversignifying breaks separating structures or cutting across a single structure. A rhizome may be broken, shattered at a given spot, but it will start up again on one of its old lines, or on new lines. You can never get rid of ants because they form an animal rhizome that can rebound time and again after most of it has been destroyed. … 5 and 6. Principle of cartography and decalcomania: a rhizome is not amenable to any structural or generative model. … [The map] is itself a part of the rhizome. …” (ATP 7-13). Manuel DeLanda proposes the term ‘meshworks’ for such systems of flat, heterogeneous, interconnectivity, which he opposes to (comparatively rigid and homogeneous) ‘hierarchies’.
 A network, in what is by now the overwhelmingly dominant sense of this term, is by definition robust. It is constructed in such a way as to tolerate massive disruption. Thus the perennial relevance of the military roots of the Internet. In its practical application, communications resilience has been inseparable from resistance to censorship. Like the basic Internet protocols, Tor (‘the onion router’) is decended from a military research program, initiated in the mid-1990s by the United States Naval Research Laboratory and subsequently pursued by DARPA. The regularity with which critical elements of state security apparatus are deflected into crypto-anarchist streams, and inversely, suggests a principle of strategic reversibility, in conformity (on one side) with the ‘poacher-turned-gamekeeper’ phenomenon. Given the prominence of treachery within all rigorously-constructed games, it should not be surprising to encounter this extreme fluidity of partisan identities, in which allies and enemies switch positions. The figure of ‘the hacker’ is notably liminal in this regard, representing – simultaneously – an ultramodern incarnation of the outlaw, and a skill-set indispensable to effective formations of contemporary power. At its highest levels, strategy is a matter of ‘turning’. This is the diplomacy intrinsic to strategy (and thus to war), once the latter is liberated from the strait-jacket of its political – or Clausewitzean – determination, and released into a conceptual space bounded only by a transcendental horizon. Friend and foe are defined within the great game, rather than outside, by its transcendent (political) frame. There can be no assumed parties at the transcendental level of the game, where not only every alignment, but any constitution of agency is itself a ‘move’. Agencies are pieces, consequent to strategic decisions. However tempting it might be to dismiss such considerations as arcane, they are already being condensed by commercial crypto-systems as practical problems. In regards to their legal form, modern business corporations have been synthetic agencies for well over a century. In the emerging epoch of Digital Autonomous Organizations (‘DAOs’), whose economic and military implications are yet scarcely glimpsed, this status advances from a matter of legal definition to one of software implementation. It consummates the historical triumph of code. At the end, as – invisibly – from the beginning, the transcendental subject is fully immanent to a plane of strategic production.