Genealogy of Algorithms: Datafication as Transvaluation

This article investigates religious ideals persistent in the datafication of information society. Its nodal point is Thomas Bayes, after whom Laplace names the primal probability algorithm. It reconsiders their mathematical innovations with Laplace's providential deism and Bayes' singular theological treatise. Conceptions of divine justice one finds among probability theorists play no small part in the algorithmic data-mining and microtargeting of Cambridge Analytica. Theological traces within mathematical computation are emphasized as the vantage over large numbers shifts to weights beyond enumeration in probability theory. Collateral secularizations of predestination and theodicy emerge as probability optimizes into Bayesian prediction and machine learning. The paper revisits the semiotics and theism of Peirce and a given beyond the probable in Whitehead to recontextualize the critiques of providence by Agamben and Foucault. It reconsiders datafication problems alongside Nietzschean valuations. Religiosity likely remains encoded within the very algorithms presumed purified by technoscientific secularity or mathematical dispassion.

and hence open to greater numbers. In a similar vein, Habermas admires Barth's "attempt to do justice to the inner normative logic of revealed faith [against] the privatization of faith." 11 Barthian de-privatizations destabilize the secret, internal, and psycho-personalized modes of Protestant belief indicted by the Weberian critique utilized into econo-rationality. Following Foucault's and Habermas's interests in Barth, critical inquiry might benefit revisiting his theology as it might apply to algorithmic probability. 12 What is at stake? If through the course of inquiry genealogy finds itself drawn into re-weighing the "Weberian set of problems of economic history and sociology," 13 then Protestant proclivities ought not to be dismissed in analyses of neoliberal power techniques. As Foucault writes: If we believe Weber, it would seem that the enrichment of an individual in sixteenth century Protestant Germany was a sign of God's arbitrary election of that individual […]. In twentieth century Germany, an individual's enrichment will be a sign […] not, of course, of God's election, [but] the daily sign of the adherence of individuals to the state. 14 Even as god disappears through secularizations of religious behavior into twentieth century economic practices, it is yet worth considering that this electoral regime remains in its transformation to state adherence. Weber suggests something even more peculiar to Protestant culture than its econo-rationality when he writes: "Even more are parliaments of periodically elected representatives […] peculiar to us." 15 [one's] own but as being in fellowship." Karl Barth, Church Dogmatics: A Selection, ed. and trans. G.W.
Bromiley (New York: Harper Torchbooks, 1961), 213;[CD 3.4]. 11 Jürgen Habermas, Between Naturalism and Religion, trans. Ciaran Cronin (Malden: Polity, 2008), 237. 12 They are not the only recent critical thinkers who share this interest. See also Jacques Derrida, "Deconstruction in America," trans. James Creech, Critical Exchange 17 (Winter 1985), 1-32; esp. 12. 13 Michel Foucault, The Birth of Biopolitics, trans. Graham Burchell (New York: Picador, 2008), 166. 14  Protestant electoral regimes remain peculiar. Today the entanglements between providence, algorithms, predestination, and probability reemerge in the digital politics of information society through this mode of belief by microtargeting. 16 It has become customary in overviews of microtargeting to focus on its successes in the 2008 Obama campaign. 17 But the technique is statistically conceived and algorithmically developed for political deployment five years prior by Alexander Gage, a Michigan market researcher. The first political iteration of the word "MicroTargeting" 18 appears in a PowerPoint slide Gage presented in 2003 to Matthew Dowd, a polling operator plotting an electoral-college strategy for George W. Bush's presidential reelection bid. 19 In earlier years of development in Michigan, Gage's larval method aided the gubernatorial campaign of George Romney, ex-Mormon missionary to the UK and father of Mitt. 20 Michigan voters do not register with parties. 21 All votes are assumed independent, which amounts to a "fatal math for Republican candidates" (which should be considered a fatal context for developments of electoral microtargeting) to partner with other rostered datasets. In Michigan this results specifically in data derived from the NRA and Christian Coalition. 22 The technique's developmental prehistory (as well as its growing datasets, patterns, and correlations) also includes Gage's earlier partnering with an Arkansas data vendor, Acxiom, whose datasets were conditioned by a "more reliant client than [even] parties and campaigns," the American Bible Society. 23 One of the largest psychographic clusters (nearly a half million) of Acxiom's data segments (once paired with the RNC voter records and applied 16 Cf. the political theology of targeting in Samuel Weber, Targets 20 Issenberg, The Victory Lab, 131-2. It should not go overlooked that this primal adoption of microtargeting into electoral politics, George "Romney's technocratic approach," is made by a gubernator characteristic of the most Americanized articulation of predestination: manifest destiny and its implicit colonialism internalized as Mormon religious calling. Cf. Weber,The Protestant Ethic,235,fn. 25. 21 State of Michigan Bureau of Elections, Lansing, "Questions and Answers: Michigan's Presidential Primary," March, 2016, https://www.michigan.gov/documents/sos/Voters_QA_MIPresPrim_516112_7.pdf. 22 Issenberg,The Victory Lab,114. nationally by Gage) was named "Bible Believers." 24 Benefitting from a base emotionally invested in a cultural war on terror, Bush's 2004 reelection took specific aim at the evangelical voting bloc. 25 It is important to maintain that the evangelical vote is a statistical minority and should not be over-estimated. 26 But is this not precisely the Protestant genius?
Critical inquiry must keep in mind these contextual kinds of data clusters and the inherited foundational traits inherent to them that, then, further compound, correlate, and optimize through the course of their development from Arkansas to Michigan to the Bush White House. All this is years in the making before microtargeting methods and techniques are adopted in fuller development by the hyperpersonalization tactics of the online Obama campaigns and eventually Cambridge Analytica (CA). Before CA was created, the test case and proof of concept the SCL Group pitched to Bannon and the Mercer family was performed in Virginia, which "has an enormous bloc of evangelical Christian voters," after buying access to their information through "Acxiom, and niche firms with specialist lists from evangelical churches." 27 It was the successful microtargeting of this particular religious electorate that secured the Mercer investment that instituted CA.
It is no secret that adherence to a simple majority fails to decide recent elections.
Scales are tipped by a mere few to a consensus or plurality. A slight yet decisive few value-up the increasingly customary split-decision between two-party systems. Only scant numbers are needed, merely a micron. The microtargeting deployed by CA exploits personal fears and is highly effective with those predisposed to believing themselves chosen. It targets "beliefs and religiosity," specifically "whether [voters] believe they control their own destiny." 28 These enable CA to target religious belief in "the just-world hypothesis" (JWH): 24 Kaiser,Targeted,85,137,175;Wylie,Mindf*ck,49. See also Vassilis Saroglou, "Religion and the Five Factors of Personality," Personality and Individual Differences 32, no. 1 (2002): 15-25. 29 Wylie,Mindf*ck, The suffix of theo-dicy derives from the Greek word for justice, dikē. The word is attributed to the title of a work by Leibniz, itself a primal text in the development of probability calculus. Noteworthy is his development of probabilistic weights of more (>) or less (<) from the gospels: "there will be many that are called and few that are chosen" (Matthew 22:14) and Paul of Tarsus "where sin abounded, grace did much more" (Romans 5:20 under-appreciated. In Augustine he finds a thinker for whom providence and predestination are not yet divided. The latter eventually becomes subordinate to the former in Calvinism. 33 This is a rare point of convergence shared between Barth and Pascal. 34 The development of probability theory that Ian Hacking, for example, suggests in Pascal is but one valence of Pascal's endeavor to distinguish Calvinist predestination from Augustine.
How could things have been otherwise? Barth revaluates predestination as grace granted prior 35 to providence. This anticipates the 'given' prior to probability suggested by Whitehead, discussed in more detail below. Barth  These latter two policies are articulated in probability theory from its inception.
Before citizens were datafied and algorithms deployed to inform or misinform microtargets, the political ambition of probability was to minimize biased passions and misinformed opinion. Laplace endeavors to apply Bayes' theorem to "decisions of assemblies," which depend not only on "the plurality of votes [but] the impartiality of the" 40 voter (secularizing the absolute indifference of a Bayesian god). Laplace's 33 Karl Barth, Church Dogmatics, vol. 2, part 2, trans. G.W. Bromiley, et. al. (Edinburgh: T&T Clark, 1957), 46, 60. 34 Blaise Pascal, Pensées and Other Writings, trans. Honor Levi (Oxford: Oxford University Press, 2008), 215-26. 35 Barth,Church Dogmatics,2.2,29. 36 Barth,Church Dogmatics,2.2,18, Barth,Church Dogmatics,2.2,11,29,71,91. 38 Only election "and not […] accompanying non-election or rejection." Barth,Church Dogmatics,2.2,16,; against the "blind fate" and "fatal parallelism" of "double predestination" (both election and rejection) in Calvin,Luther,and Zwingli. Barth,Church Dogmatics,2.2,17,25. 39  Bayes arguably offers a revaluation of Calvin centuries prior to Barth's genealogy. This is a doctrine of election "not" by "only" a chosen number rejecting greater or lesser remains, but rather a layering of ranked numerical degrees. One might even detect here-prior to Laplace-a schematic tabulation of an actual ranked-voting ballot (or at least a collateral tabulation of its results) initiated by the namesake of the primal probability algorithm, 50 himself, conceptualized within his singular providential theology informed by ground-breaking mathematics. Activist advancement for ranked-choice voting is merely one possibility by which genealogical inquiry into the political theology of providence, probability, and algorithms might be practiced in hopes of improving public life with amendable revaluations of the promise of large numbers. In Whiteheadian terms, it strains to value-down oppositional, partisan, private, or personal biases rather than value-up those of a microtargeted chosen few… for the enrichment of far fewer.

Shay < Column > Icon
The previous section suggests ways by which electoral politics in datafied information society can exploit religious convictions (theodicy and predestination) by way of algorithmic computation (data-mining and Cambridge Analytica's probabilistic simulation/prediction of electoral behavior patterns). This section further inquires into the roles of religion and theology within the mathematical development of algorithms. The tenets of admixture by Zellini stated at the beginning of this article are adopted by the Italian biologist and mathematician, Giuseppe Longo. Longo maintains Zellini's position that mathematics is "shaped by metaphysical debate" from Euclid and Aristotle to Aquinas. Critical inquiry must ask, along with Longo, "how [was metaphysics] picked up by Mathematics" as a discursive object and proof component? 51 Longo notes two essential innovations made by theology in the development of mathematical notation, specifically Islamic theoeconomics at the time of al Khwarizmi's innovations (of the Hindu art of reckoning) and the subsequent aesthetic geometry of Christendom. 50 Bayes' theorem, discussed in more detail below. The "nature of God's knowledge among early Muslim theologians [e.g., Zaydî Sulaymân] led to the assumption that it was a 'thing' (shay')." 54 Many of these theologians "with Christian antecedents […] merely transferred their [Christian] accustomed patterns of thought to the new religion." 55 An important example is effectiveness, 56 how pondering the "effect (athar)" of a thing is considered an "inferior mode of perception inadmissible in God." 57 The formulaic elements that give rise to the algorithm enact civic principles here-below, according to Qu'ranic instruction.
Al Khwarizmi formulated a specifically verbal practice (or rhetorical algebra) that grows from the first Arabic dictionary combining the "'divine language' (Quran)" 52 Rosenthal,Knowledge Triumphant,112,cf. 98. with "pagan" Arabic spoken prior to Islam, based on phonemes that combine into "words with neither phonetic or semantic value," 58 "drawing not at all on symbolism, but only on terms from natural language." "The algebraists [al Khalīl and al Khwarizmi] adapt themselves to this kind of relationship as well." 59 Algebra is born as an operation of "restoring [or] reduction," "rectifying or correcting using some form of constraint" emerging as "a discipline no one had ever conceived [in] which one can indifferently reduce both arithmetic and geometrical problems." 60 Algorithmization by shays is initially formulated to resist idolatrous tendencies to the dead letter of the law-written symbology-indicative of Abrahamic religions (and monotheisms more broadly). It aspires to iconicity against potential misuse as idolatry. The algebraic capacity to find correlations beyond individuated shay punctuality applied to correlates learnable by human machines, but nevertheless correlated to truths designated distinctively to divine reckoning. Islami and Longo make little direct reference to the Qu'ran. Noteworthy is a divine practice of metic- infinity into the finite." 64 Focusing on the paintings of Lorenzetti (in which a christic "column" attenuates to-and overlaps with-a vanishing axis), there emerges "an extraordinary innovation: a rigorously drawn projective space, with a limit line, not just a point […]. Then, by the effect of the geometry […] that goes from man to God, a new space is deployed." 65 The new space deployed from this inventive column 66 is a monumental precursor in the history of mathematics that develops into the "epistopic viewpoint" that can be operated within vector space, which "generates a new column-vector […] and some new numbers, statistics." 67 This mathematical infinity through mariological imaging "is a tool for the intelligibility of the world." 68 Even if aesthetics makes for unviable politics, it has nevertheless contributed to "a buildable world." 69 In Renaissance painting, "projective geometry, a mystical decision, organizes at once the space of God and of man. One discerns the legacy of thing semblance by algorithmic shays in Peirce's icon.
"Anything […] is an Icon of anything, [but only] insofar as it is like that thing and used as a sign of it." 82 Shayed rhetorical algebra recalls Peirce's description of interpretant symbols that "grow" from multiple "icons" 83 as "pure rhetoric." 84 Al Khwarizmi's shay is perhaps an icon that is not yet a sign. But as it operates (or is simply thought) as an element of a composite algorithm, it becomes iconic: "[E]very algebraical equation is an icon." 85 The qualitative distinction between Islamic law and its notation is maintainable in Peirce since an icon refers to its object, "whether any such Object actually exists or not." 86 This arguably applies as much to the non-ontological status of transcendental numbers such as p as to a rhetorical balancing of algebraic numbers endeavored by al Khwarizmi. By a process "like mathematical reasoning, we can reach conclusions as to what would be true signs [by which] modes of thought of a God, who should possess an intuitive omniscience superseding reason, are put out of the question." 87 79 Mackenzie, Machine Learners, 71; emphasis added. 80 Mackenzie,Machine Learners,91. 81 The "curse of dimensionality" described by Richard Bellman, discussed in Mackenzie, Machine Learners, 62, fn. 6. Q.v., 111, fn. 8 (on k nearest neighbors). 82 Peirce, "Logic as Semiotic: The Theory of Signs," Philosophical Writings of Peirce, 102. 83 Peirce, Philosophical Writings of Peirce, 115; the plurality is crucial. 84 Peirce, Philosophical Writings of Peirce, 99. 85 Peirce, Philosophical Writings of Peirce, 107. 86 Peirce, Philosophical Writings of Peirce,102. It would be a mistake to reductively interpret this as simple scientific dismissal of god by mathematics. It might be considered alongside the theoeconomic politics of al Khwarizmi and algorithmic implementation of Islamic law. In his explanation of a symbol as legisign (a law that is a sign; a sign that refers to its denoted object "by virtue of a law"), Peirce writes that "this law is usually established by men." 88 Such mathematical reasoning might describe how the function of a shay, icon, or algorithm is a pragmatic practice of delimited human reason applied to creaturely existence distinct from whatever those elements may mean to divine reason (if anything, and, as such, resists idolizing its own capacities as anything resembling conceptions of gods). One discerns influences of both al Khwarizmi's algorithmic ambition against idolatry and expectant human judgments in Bayesian Divine Benevolence as Peirce explains how "discoveries of science [and] their enabling us to predict what will be the course of nature, is proof conclusive that, though we cannot think any thought of God's, we can catch a fragment of [God's] Thought, as it were." 89 The shay (and arguably its algorithm) is such a fragment and intends nothing more beyond.
Peirce is outspoken and confessional about his theism. 90 He is a probability theorist 91 as much as he is also a Pauline thinker. His essays, "On the Doctrine of Chances, with Later Reflections" and, in particular, "Evolutionary Love," evince a commitment to salvage both Paul 92 and the gospels 93 from the opportunistic theoeconomic political theology at the basis of the Hellenization of their messianic tradition of love 94 into an institutionalized Roman state religion 95 and its subsequent perversion into the Protestant ethic of capitalism. 96 These commitments are not divorced from his probability theory. He takes recourse to both "the principles of probability" 97 and empiri-Brower: Genealogy of Algorithms 17 cal sociological statistical data in the very same discussion. Bayes similarly discusses the rift among mathematicians regarding proportion and ratio in Divine Benevolence. 98 Peirce's probability is nearly apostolic as a pragmatic philosophy of love through which he explains even Darwin with energetic agapism. 99 Peirce is no doubt participant in Protestant proclivities, but he is not uncritical of these tendencies as well as of Hegel, from whom he inherits them. Peirce practices a probabilistic logic poignantly critical of both the logic of predestination 100 and its collateral complacent physics reduced to blind mechanistic determinism (tychasm) 101 which he finds scathingly unscientific, obstructive to inquiry, 102 and apostate to the true promise of probability calculus.
Peirce's concern with god attempts to distinguish itself from religion. In "The Approach to Metaphysics" he suggests scientific curiosity much analogous to theology and regrets the religious tendencies of theologians to devolve into an "army [of] sworn fidelity." 103 The Anglo-Hegelian tendency identified by James, as well as this irreligious, yet theistic, trend in probability theory persists in the works of Alfred North Whitehead. The next section attempts to attend to ways by which these trends emerge within Whitehead's criticism of probability.

Given Contingents
This section follows a certain "Whitehead revival" 104 in recent scholarship. It is discernible in thinkers inspired by Foucault 100 Peirce, Philosophical Writings of Peirce, 371; Cf. the "faith of the logician" in "certain predetermined conclusions" in "What is a Leading Principle?," in Philosophical Writings of Peirce,130. 101 Peirce, Philosophical Writings of Peirce, 364-70, which is specifically Constantinian (368). 102 Peirce, "The Concept of God," in Philosophical Writings of Peirce, 377. 103 Peirce, Philosophical Writings of Peirce, 313. Cf. the difference between "ecclesial reasons" and "theological ones" in Giorgio Agamben Whitehead exemplifies a mathematical mastery contributing to probability theory at a developmental pinnacle, while also levelling criticisms.
For Whitehead, Laplacean physics of efficient causation embodies a potentiality that is yet "the correlative of [a prior] 'givenness.' The meaning of 'givenness' is that what is 'given' might not have been 'given'; and what is not 'given' might have been 'given.'" 107 There must "be limits to the claim that all the elements in the universe are explicable by 'theory.' For 'theory' itself requires that there be 'given' elements so as to form the material for theorizing." 108 Probability is no exception. Its statistical application relies on a given. The given is that which gives activation to any mathematical formulation of it. Yet the given seems ever prone to escape the presumptive attempts by statistical probability to grasp it as foundational to practical application.
Chapter 9, §5 of Process and Reality questions the legitimacy of probability by its own self-referential measure.
The notion of 'probability,' in the widest sense of that term, presents a puz- Mathematical application remains probability's proper horizon. Whitehead claims that any "ultimate ground" to equiprobability (e.g., that any of the six sides of a die may land face up, all six being equally probable) must be "explicable without reference to any notion of probability." 112 The six equiprobabilities must be as distinct from probability itself as the six respective faces of the die are distinct from the die itself. "A die is a given fact; and its faces do not differ […] in any circumstance relative to their fall […] beyond this given fact, there is ignorance." 113 But since the ultimate ground for Whitehead "must not require any appeal to probability beyond itself […] the statistical facts as to the ground, must be 'given' and not merely 'probable.'" 114 That which lies beyond probability must be given to it from beyond the scope of its proper vantage, but on which it nonetheless relies. Statistical probability seems unable to quantify or enumerate an element necessarily extrinsic to it that is yet required, according to Whitehead, if it is to be more than merely probable. 109

Brower: Genealogy of Algorithms 20
A given, understood as "not merely 'probable,'" invites alternative considerations of datafication that might be worth trying to salvage (akin to the Nietzschean ambition towards revaluation). The "alternative non-statistical ground" 115 Whitehead offers is an attempted revaluation of data as properly given: data worthy of the name. This would be a strained reception of one's self-datafication by the given world of one's experiential surroundings. Once Whitehead discloses the given non-statistical ground for probability, he further socializes the given as data of the world one experiences. Probability and the given converge in data received by a judging subject: "[T]hese data are not extrinsic to the [subjective] entity; they constitute that display of the universe which is inherent in the entity. Thus the data upon which the subject passes judgment are themselves components conditioning the character of the judging subject." 116 The process occurs between "every creature" and the order "constituting the primordial nature of God." 117 Whitehead defines the latter as god's "complete envisagement of eternal objects." 118 A more socialized mode of datafication emerges through a qualitative distinction from which subjective judgements are not merely datafied by the physical or statistical data of vulgar temporality. Such "objective data" 119 discover merely "causal feelings" 120 that do not merit valuation at all. But the "subjective form of a conceptual feeling has the character of a 'valuation'" 121 that "arises [as] the eternal object, which is the datum of the conceptual feeling, is an ingredient in some sort of datum in which the other components are the objective data […]. This new datum is the integrated datum […], some sort of contrast." 122 Conceptually datafied, as such, subjective judgment of eternal objects evaluates by ascension or descension: "[T]he valuation of the conceptual feeling is a 'valuation up' or a 'valuation down,' 115 Whitehead, Process and Reality, 314. 116 Whitehead, Process and Reality, 309. 117 Whitehead, Process and Reality, 315. 118 Whitehead,Process and Reality,70. 119 Whitehead,Process and Reality,367,402,472. 120 Whitehead, Process and Reality, 361, 365-6. 121 Whitehead, Process and Reality, 367.
Brower: Genealogy of Algorithms 21 the importance [therefore is either] enhanced, or attenuated." 123 This rearticulates the para-numerical "addition and diminution" 124 that Bayes attributes to god and remains crucial to probability theory. 125 Whitehead's "given" gradually grows into a god of process theology nearly synonymous with his very name. Reminiscent of Naturreligion (no doubt influenced by Hegel), Whitehead suggests it is beyond categorization as religion. He is aware that he is performing a secularization of theological concepts. He concedes this and even suggests that "the secularization of the concept of God's functions" is a necessary, "urgent […] requisite of [human] thought." 126 He intimates an important distinction between theology and religion: "It must not be assumed that my non-statistical judgments are in any sense religious." 127 Whereas the "concept of god" may be an "essential element in religious feeling," the "converse is not" the case: "[R]eligious feeling is not an essential element in the concept of God's function in the universe." 128 Whitehead evokes god without forfeiting mathematical probability, a Pascalian insight, rearticulated in Ramsey 129 and Peirce, above. Whitehead endeavors not to dismiss probability but to correct it from miscategorical tendencies to transvaluate its conditional given into just a statistically enumerated datum. Not only does his theology remain compatible with mathematical probability, it seems urgent and requisite that probability take account of itself as secularized theology. By falsely grounding itself on its own efficacy, which is no doubt remarkable (yet nevertheless difficult to dissociate from the self-sufficient ontotheologies of Anselm, Descartes, or Hegel), probability seems prone to mutate into a secular fundamentalism akin to the very religiosity it presumes to overcome.
To give a brief sketch, the remaining sections of this article attempt to employ key motifs in Whitehead's critique of probability as they apply to three exemplary 123 Whitehead,Process and Reality,368. 124 Dale, The Life and Work of Thomas Bayes,117. 125 John Maynard Keynes, A Treatise on Probability (Lexington: Wildside Press, 2017), 20-3. 126 Whitehead, Process and Reality, 315. 127 Whitehead, Process and Reality, 315. 128 Whitehead, Process and Reality, 315-6. Science and the Modern World devotes more patience to religion. practitioners of genealogy: Foucault, Agamben, and Nietzsche. It will return to the latter keeping in mind a given beyond the merely probable. But these next two sections address Foucault and Agamben attending to problems of political theology alongside evaluative envisagement. 130

Theodicy & Theogony
The envisagement of Whitehead converges with a supra-historical perspective overviewing from a site of truth described in Foucault's "Nietzsche, Genealogy, History." The opening section, above, suggested ways by which the algorithmic microtargeting of Cambridge Analytica deployed information tailored to recipients whose statistical psychography identified a significant probability of religious interpretations of the just world hypothesis. In the political aftermath of its efficacy, it is perhaps not surprising to recall that theodicy has always been one of genealogy's primary targets.
Foucault suggests that it seems prone to inhibit any genealogical thinking at all.
As he explains that no "genealogy of values" should ever be confused "with a quest for their 'origins," 131 Foucault returns to the preface of the Genealogy of Morals as Nietzsche retraces his own primitive concerns with theodicy: "involvement with the question: […] if God must be held responsible for the origin of evil. [Nietzsche] now finds this question amusing and properly characterizes it as a search for Ursprung." 132 Foucault discloses an entanglement between primal theodicy and theogony. This "lofty origin […] is associated with the gods, and its story is always sung as a theogony." 133 But the genealogical overcoming of outdated theogony is perhaps premature.
Given time, the emergence of probability ushers in a reemergence of this strange 130 Importantly, this is only one of several valences attributable to the term 'envisagement' in Whitehead. Further ruminations on these themes could climax in a naturalistic theogony that would study the structure of this hierarchy […]. 136 Foucault suggests that Ursprung fixation fabricates otherworldly religiosities because it is prone to think effects as causes. 137 It errs by "complete reversals" 138 or inversions: 134 Nick Bostrom, Superintelligence (Oxford: Oxford University Press, 2014), 12-3. 135 Foucault, "Nietzsche, Genealogy, History," 82. It is difficult to imagine a more explicit codification of causal reversal than P(effect | cause). Its reversibility is precisely Domingos' intention: [W]hat we usually know is the probability of the effects given the cause Its statistical application reverse-engineers from given data collected by past empirical experience to predict probabilities of future events, to infer inversely from effects observed to causes of their probable repetitions to come. The "devotion [of Bayesians] to this idea is near religious." 153 Bayes is not the first to think from effect to cause. He is a Protestant parrot.
Calvin rarely gets the credit he deserves for this contribution to intellectual history.
The Institutes introduce a new causal hybridity (primacy, efficiency, and love), conceptual only from divine vantage to which human works are not the cause of election since the efficient cause of election is divine love. This anticipates Peirce's recourse to primal agapism while endeavoring to correct probability from both religious predestination and deterministic scientism, discussed above. is in no way dependent on creaturely works or effects. By "observing the works of God," one "will always observe the marks of goodness" that "are the effects of general laws." 155 Bayes stresses the importance of repeated observation and expectation. Through observation (the "greater number" of which "we know," provided the observed events are "really distinct from each other"), allows observers to "judge what […] to expect from Him [god]." 156 In such "effects of general laws […] we shall never find any that really overthrow our notion of a perfectly good and benevolent Deity." 157 From the divine vantage, any unilateral successive order from cause to effect seems merely a human shortcoming: "[T]here is no manner of reason to suppose, that any particular order or proportion of things appears to the divine mind more excellent […] than another." 158 My opening sections focused on entanglements of Protestantism with politics, providence, and probability. The previous section emphasized Whitehead's critique of statistical deployments of probability by way of a given beyond the probable and the envisagement of god. This section closes upon consideration that Calvin and Bayes are perhaps complicit in the kind of reversal of cause and effect, which Foucault suggests originates in the implicit theogony assumed from a supra-historical perspective and indissociable with theodicy. We move into the next section anticipating that, as Agamben attempts to further develop Foucault's critique, he notes the special perspective of governance and providence. This develops into his own confrontation with probability. Agamben notes that Pascal insists on an interruption of a game for the sake of predictive calculation. This demands a state of suspension, deploying governmentality through which the game of politics is revealed: "This means that probability is never punctually realized [but] it allows us to intervene in reality, as considered 167 Agamben, The Kingdom and the Glory, 86. 168 Foucault, "Nietzsche, Genealogy, History," 88-9. 169 Foucault, "Nietzsche, Genealogy, History," 92. 170 Agamben, What is Real?, 17. 171 Hacking, The Emergence of Probability, 2. The effectual ontology of gubernatio in The Kingdom and the Glory converges with the "ontology of command" 177 in Opus Dei and supplements "the ontology of the probable" 178 in What Is Real? The "area in which" calculation and events "overlap" 179 173 Agamben, What is Real?, 34-5; emphasis added. 174 Agamben, What is Real?, 12-13; emphasis added. 175 Agamben,What Is Real?,28. is the vectoral cyberspace where algorithms function as machine-learning via vast data mining at a site of chosen perspective from which to compute correlations and formulate probabilistic predictions.

Providence & Probability
As humans initially create god in their own image, they come to operate and encode algorithms in god's image. This is the political theology of probability and the secularized theodicy of machine learning. Theodicy had claimed that only god's special perspective might discern the overall glory and benevolence of creation (imperceptible or incomprehensible to the stunted perspective of temporal creatures). This becomes the popular operation of rationalizing evil or suffering by formulation of the just world hypothesis. Such justifications extrapolate probabilistic recreations of realities in images naively presumed to be merely human yet ever encrypted as a god.
Political government, genealogically enabled by divine gubernatio, now algorithmically self-optimizes by statistical correlations of machine learning. The very name, 'cybernetics,' of course shares these gubernatorial roots with providential theology. This is reflected in Wiener's interest in Augustine and golems. Wiener explains his coinage of the word "derived from the Greek word kubernetes, […] the same Greek word from which we eventually derive our word for 'governor.'" Only later did he learn "that the word had already been used […] with reference to political science [in] the earlier part of the nineteenth century." 180 But the theopolitical legacy of cyber/gubernation is at work long before, as analyzed in exhaustive genealogical detail by Agamben throughout The Kingdom and the Glory.

Umwertung & Datafication
If Whitehead's evaluative process, discussed above, falls short of Nietzschean revaluation, it at least endeavors to correct upward toward something akin to it.

Brower: Genealogy of Algorithms 32
Today the human machine is dangerously displaced and supplemented with a "digital doppelgänger," 181 "quantified self," 182 or "digitized identity" 183 compounded into "algorithmic identities" 184 conflated into "datafied selves." 185 The subtle but solemn differences between identity and self take on new importance in the information age. These technical forms of "transindividuality" 186 undergo a transvaluation of the very meaning of data. As Nietzsche unveils values inversely coopted to misrepresent any originary value (if there is such a thing), today data suffers such reversal.
Data is devalued as this Umwertung virtually datafies anything that might have been valuable. Philology reminds us what seems increasingly forgotten or repressed: that the term "data" comes from the Latin word datum and means "given." Not long ago, data invoked and denoted givens. But this is no longer the case. As Nowotny writes: "Welcome to the age of big data. Despite their name, they are no longer given, but made." 187 In times of crisis, datafication machines even attest to this transvaluation of alogy already suggest that problems of given-data might be even more primal than those of value itself. The two are likely indissociable. This is perhaps why data seems prone to valuation or transvaluation might become predisposed to appropriate the data of any given time. It is perhaps because given-data are prone to misappropriation that values are, collaterally, apt to be transvaluated. Genealogy must suspect tacit transdatafications within the vast accumulation and refinement of big data.
Statistics also gives way to value inversions: " [C]ontrary to what the etymology of the unfortunate term 'data' suggests, very few 'data' are actually 'given. '" 190 As data becomes more utilized as statistics, its socioeconomic transvaluations assume statistical articulations. Like the long forgotten language in which the referent of "data" remains "given," the more publicized "value added" (of national taxation and accounting) in statistical languages of the past "appears to have been forgotten" as statistical languages progress into more personalized "value creation" (of finance and assets of shareholders). 191 These emerge through financial formulations of privatization by which any "socialist" concern can no longer integrate its objectives into an ordoliberal "political game, since the givens had been reversed. This section suggests that Nietzsche's introduction of the dangers of transvaluation by way of a primal problematization of values given as factual might perhaps still have much to teach us about the current datafication operations proliferating our everyday lives. It is worth considering that the "new datum" of valuation articulated by Whitehead, above, discloses possible datafication processes worth embracing in critical resistance to the transvaluative tendencies formulated through statistical probability, big data, or surveillance and platform capitalisms. 198 If contemporary 195 Nowotny, The Cunning of Uncertainty, 41-2. 196 Nowotny, The Cunning of Uncertainty, 42. 197 Koopman, How We Became Our Data, 14, 42, 35-107, 171, 189. 198 Nick Srnicek, Platform Capitalism (Malden: Polity, 2017), 57. datafication mechanisms are re-emerging through digitized activations or computational deployments of transvaluation, then there is all the more reason to reconsider the possibility by which Whitehead's understanding of a given beyond the merely probable might enable revaluations of data beyond the merely enumerable. This would seem to require a reconsideration of the secularization of theological concepts beyond the merely religious, especially if the more religious we are the more merely probable we shall likely remain. 199 But a possible alliance survives between any Nietzschean politics to come with a Laplacean electoral practice driven to minimize ressentiment with probability, which yet remains unapologetically influenced by providential deism saturated in an alleged divine indifference.

Conclusion
The epigraph from the editors of WIRED on which this article commenced might be forgiven for heralding inevitable reversals encoded into computational platforms.
But as genealogy works its way from Cambridge Analytica back to Nietzsche, this could only ever be considered inevitable if we have already succumbed to the alleged reversals without critically suspecting them of transvaluating through digital disguises of datafication. We began on a possible admixture of religion and algorithmic calculation. I suggested a continuation of this into further admixtures of Protestant predestination and algorithmic probability through manipulative microtargeting.
This tactic develops as much from statistical probability as from psychographic information correlations to specifically religious data clusters. The probability of Laplace and Bayes are arguably already critically attuned to such problems. Electoral techniques that value-up smaller numbers to determine an allegedly larger consensus are statistical redeployments of the chosen few characteristic of Calvinist predestination (itself a redeployment of an Abrahamic chosen people). Even the appealing alternative of ranked ballot voting aspires to a Laplacean impartiality on the part of the voter that is yet a subjective secularization of divine indifference presumed in the providential theology of Bayes. 199 Cf. the second epigraph above from Paley, Natural Theology.
Such indifference is presumed indifferent to creaturely effects. A Christian conception of effectiveness survives the Abrahamic transition to Islam, the theopolitics and legalism of which are indissociable from the advent of algebra and algorithms as articulated by al Khwarizmi. The innovation of projective geometry Longo finds in religious paintings sets the stage for the statistical correlations made operational from the perspective of a chosen point in the virtual dimensions of vectoral cyberspace made visibile to machine learning, critically explained by Mackenzie. Mathematical, scientific, and pragmatist critiques can be found that all insist on the importance of not dismissing the theistic or religious undertones that develop over time into modern algorithmic techniques. The pragmatist probability of Peirce articulates itself in collateral critiques of religious predestination and scientistic determinism. This is echoed by Whitehead insisting on a non-statistical givenness prior to and beyond the merely probable. I suggest that this is a mathematical iteration of the electing grace prior to providence in Barth. The nearly Nietzschean valuations enabled by the given in Whitehead are founded on the envisagement of god. Mathematical probability, as such, is equally enabled to move beyond any chosen religion without forfeiting the concept of god. The more the concept of god is dismissed to the whims of religion, the more religious sentiments remain exploitable as microtargets prone to believing themselves chosen.
Religious conceptions of envisagement assume the supra-historical perspective critiqued by Foucault in terms of providence and theogony. These function through a reversal of causation associated with religious conceptions of an afterlife. But causal reversal must also be considered at work in both the mathematical probability calculus of Bayes' theorem derived from the Essay on Chances and the providential theology of Bayes' theodicy in Divine Benevolence. Neither is indemnified from the other. To this day Bayes' theorem remains the core code of contemporary algorithms, be they deployed as sorting predictions by the Google search engine or microtargeted advertisements by Amazon. Agamben furthers the Foucauldian critique of governmentality through probability's reliance on a special perspective generalized in order to govern. The earlier admixture of predestination and election problematized by Barth and Pascal anticipates the admixture of gubernation and providence in Agamben.
This contextualizes not only the governmental capacities of cybernetics, but even the very word coined by Wiener. If there is a given reality beyond the merely probable as Whitehead suggests, this in no way incapacitates the possible activation of algorithmic probability to interrupt, control, or command over it. The spirit of capitalism indicted in the Weberian critique still spooks our valuations. Wiener always regretted his inability to communicate the application of cybernetics and probability to the benefit of "labor unions." 200 He confesses his eventual recourse to psychological and physiological research as a form of ascetic retreat "most remote from war and exploitation" 201 against the tendency to minimize "human values" merely to maximize "buying and selling." 202 Just as there is no human lifeform that is not to some extent cyborg, there is no algorithm indemnified from some element of humanity. Even base-10 decimal enumeration is a byproduct of aleatory evolution into the anthropoid manual anatomy of fingers and thumbs. 203 But it seems incumbent for any human element, as such, to attend to what Peirce considers a precious "instinct" 204 effectuated by anthropogenesis toward theism, rather than feign victory over it by assuagement or repression.
There are no de facto malicious algorithms or data practices, only formulaic activations of them by an equally human element of self-interest, biased desire, fear, or calling: the very opinions Laplace longed for probability to minimize.
It has become commonplace in contemporary critical circles, even in seminaries, to identify the death of god as a distinctly Christian invention. 205 It would be a precious cultural accomplishment to the Christian tradition's credit if this were the case. Genealogical inquiry suggests that the same likely applies to probability theory. This includes the technical practices and algorithms by which it functions and 200 Wiener, Cybernetics, 28. Cf. Barth's organizational labor activism in Angela Dienhart Hancock, Karl operates. Cursory awareness of the Bayesian legacy set forth from the Presbyterian pulpit might easily identify probability as the new religion and its administrative algorithms as the new angels. Both remain susceptible to the seductions of religious metaphysics and must be criticized accordingly.

Funding Information
This work is a part of the research project "Paradoxes of Theological Turns in Contemporary Culture," Univerzita Karlova, Praha, PRIMUS/HUM/23.