Relativizing the Classical Tradition: Hartshorne’s History of God

Abstract

This paper follows Hartshorne’s thinking about the varieties of theism, especially through his use of position matricies. I pinpoint some of the limitations of this method but also expand upon it so as to remedy some of its drawbacks. One can prove, augmenting Hartshorne’s method, that there are far more concepts of God than he realized. Hartshorne’s method, moreover, highlights the fallacy of equating theism and supernaturalism and frees the imagination to view God in naturalistic terms without collapsing into atheism. At the very least, some version of theistic naturalism stands along classical theism and atheism as a live metaphysical option. Finally, one can apply Hartshorne’s thinking to the meta-level problem of religious language and thereby clarify options among various types of kataphatic and apophatic theologies.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

A theme iterated throughout the writings of Charles Hartshorne (1897–2000) is that philosophers and theologians have been insufficiently attentive to the historically significant and logically possible meanings of theism. The dominant tradition in Western theology, drawing on certain elements of ancient Greek philosophy, conceived God as an immutable and infinite perfection, capable of existing without the universe, but as freely creating it from no preexisting material, guaranteeing its continued existence, and governing its destiny. “Classical theism,” as Hartshorne called this view, became the paradigm; what theists and atheists considered themselves to disagree about was classical theism. Any deviation from the doctrine had to meet a burden of proof or be considered deficient. Thus, the alternatives to classical theism were characterized, and caricatured, as doctrines of a merely finite, limited, or imperfect God which could hardly be expected to command serious philosophical attention, much less religious devotion.

To a great extent, this attitude still prevails, as one can see in the contemporary movement called “the new atheism.” Victor Stenger refers to “highly abstracted concepts of a god” developed by sophisticated theologians that would be unrecognizable to typical believers (Stenger 2007: 112). Other new atheists suggest that alternate forms of theism are disguised forms of atheism. Recently, at the Claremont School of Theology, Daniel Dennett averred that the difference between atheism and the views of Philip Clayton, a Christian theologian who questions elements of the classical tradition, is a difference in name only. About the same time, a friend of mine, sympathetic to the views of Richard Dawkins, wondered how process theism is different from “sexed-up atheism.” These examples suggest that, where the theistic question is concerned one’s options are: classical theism, atheism, or a watered-down version of theism that is really a version of atheism. Hartshorne’s reflections on the varieties of theism pose a serious challenge to this trilemma. His approach to the theistic question engages classical theism without privileging it as the default position in philosophical theology.

Hartshorne argued that a fair assessment of theism requires an honest attempt to explore the possible meanings of the concept of God. Indeed, the question whether God exists presupposes at least a minimal understanding of what is meant by the word “God.” With the aim of finding different concepts of God, Hartshorne employed two interrelated methods: First, he studied the history of philosophy, paying special attention to theologies, both Eastern and Western, at variance with classical theism. Second, he developed conceptual tools for exploring logically exhaustive sets of options in ways of thinking about deity. Only after his ninetieth birthday was he satisfied that he had found the most perspicuous array of options for thinking about theism.

My purpose in this paper is to follow Hartshorne’s thinking about the varieties of theism, especially through his use of position matricies. I will pinpoint some of the limitations of this method but also expand upon it so as to remedy some of its drawbacks. One can prove, augmenting Hartshorne’s method, that there are far more concepts of God than he realized. Hartshorne’s method, moreover, highlights the fallacy of equating theism and supernaturalism and frees the imagination to view God in naturalistic terms without collapsing into atheism. At the very least, some version of theistic naturalism stands along classical theism and atheism as a live metaphysical option. Finally, one can apply Hartshorne’s thinking to the meta-level problem of religious language and thereby clarify options among various types of kataphatic and apophatic theologies.

The Historical Approach

Hartshorne’s interest in the variety of meanings that have been given to the concept of God is illustrated in a story that he relates about Arthur Lovejoy, the great historian of ideas.

[Lovejoy] was proposed for some community responsibility, and a political committee, interviewing him to assess his fitness for the office, asked him if he believed in God. One must know Lovejoy to take his reply in the right way. It was to the effect that, so far as he knew, a number (I think nine) of different meanings had been assigned to the word God. He proceeded to explicate these meanings. I don’t know if he ever got to the point of endorsing one of the conceptions, or rejecting them all, or declaring his inability to decide. He was given the responsibility. With some persons there might have been evasiveness or cynicism in such a procedure. But to a Lovejoy nothing is more true than that terms of philosophical and religious dispute are full of ambiguities, so that it is idle to take a position for or against without the most particular examination of the way in which key words are used (Hartshorne 1990: 320).

Hartshorne liked this story because it shows that Lovejoy understood that the question “Does God exist?” presupposes the question “What do you mean by ‘God’?” He, and Hartshorne, understood that, taking into account what intelligent and sensitive people the world over have believed, the second question has no simple answer.

Lovejoy’s insights found popular expression in Karen Armstrong’s bestselling book A History of God. Armstrong canvasses a bewildering array of perspectives on the idea of God in Western culture, from ancient Babylonian creation myths to the musings of scientists, philosophers, and theologians in the twentieth century. Despite some missteps, Armstrong’s “history” is a useful reminder of the variety of forms theism has taken.¹ What is missing, however, from her narrative is any attempt, such as Lovejoy apparently made, to provide a systematic presentation of theistic doctrines.

Hartshorne, like Armstrong, was interested in the history of the idea of God (extended to include nonwestern ideas), and, following Lovejoy, he explored various ways of mapping the logically possible varieties of theism. Hartshorne treated the history of philosophy “as a laboratory of intellectual experiments in theories, and arguments for or against theories, and in judgments about theories and arguments” (1993b: 308; cf. Hahn 1991: 633). His approach to the history of philosophy is less a history of great thinkers or great systems of thought than a history of great ideas. In this way he attempted to avoid the criticism he made of others that “minor points by great philosophers are dealt with, often with loving care, but major points by minor philosophers are missed” (1970: 86). Where the concept of God is concerned, the historical approach is most clearly illustrated in Philosophers Speak of God (1953, 2000), edited with William L. Reese. This book presents selections from the writings of 52 philosophers and theologians as well as excerpts from the scriptures of Hinduism, Buddhism, Judaism, and Christianity. What sets this anthology apart is the inclusion of philosophers, both well-known and obscure, from both Eastern and Western traditions. For example, alongside writings of Plato, Aristotle, Aquinas, Hume, and Kant are selections from Ikhnaton, Channing, Ramanuja, Iqbal, and Lequyer.

Hartshorne and Reese grouped theistic doctrines in terms of five questions: Is God eternal? Is God temporal? Is God conscious? Does God know the world? and Does God include the world? Answering any of these questions in the affirmative is symbolized with a single letter: E for eternal, T for temporal, C for conscious, K for knowing the world, and W for including the world (where “including the world” means “having all things as constituents”) (2000: 16). An affirmative answer to all the questions yields Hartshorne’s panentheism, symbolized ETCKW; Hartshorne would eventually refer to his form of theism as “neoclassical” (1962, passim). Hartshorne and Reese are clear that E and T are contraries rather than contradictories. God may be both eternal and temporal, but in different respects. For example, a God that is never born and never dies could be characterized as eternal, but that same God may experience the passage of time and thus be characterized as temporal. When E and T occur together it is understood that they do not apply to God in the same respect.

Forms of theism that diverge from ETCKW are construed by Hartshorne and Reese as partial denials, or “truncated” versions, of it (2000: 17). For example, Aristotle’s theism is EC, meaning that God is eternal and conscious, but God is not temporal, does not know the world, and does not include the world. Classical theism is ECK, and classical pantheism is ECKW. What Hartshorne and Reese call temporalistic theism, symbolized ETCK, is represented by Fausto Socinus and Jules Lequyer. Contemporary adherents of ETCK might call it “free will theism” or “open theism” (Basinger 1996; Pinnock et al. 1994). Also included in the list of alternatives are Plotinus’ emanationism (E), extreme temporalistic theism as in Samuel Alexander’s theology (TCK), Henry Nelson Wieman’s radical temporalistic theism (T), and various forms of atheism. (Hartshorne and Reese use the expression “extreme temporalistic theism” to describe both TCK and T.)

If one assumes ETCKW as a starting point, a formal analysis reveals 32 doctrinal possibilities, 23 more that Hartshorne and Reese explicitly identify. This can be seen most clearly by considering ETCKW as a set and charting a field of sets over it—atheism is here defined as the empty set (Table 1). Hartshorne and Reese consider some principles for eliminating alternatives. For example, if God cannot know the world without being conscious—if “K  →  C” is true—then one eliminates eight options—ETKW, ETK, EKW, TKW, EK, TK, KW, and K (2000: 17). Of course, the principle that consciousness is a necessary condition of knowledge would have to be debated before these eight alternatives could be eliminated. Another possible principle for the elimination of alternatives is that a being that includes the world and is conscious must know what it includes; if one assumes “(C Ù W)  →  K” one eliminates ETCW, ECW, TCW, and CW. Hartshorne and Reese also argue that every combination should speak to the temporal/eternal status of God, that is, (x)(Ex ∨ Tx); in that case, the following options must be omitted: CKW, CK, CW, KW, C, K, and W. This still leaves seven alternatives that Hartshorne and Reese do not consider: TCKW, ETC., ETW, ET, TC, EW, TW.

Table 1 ETCKW and its subsets

If some doctrines can be eliminated other options are not addressed in the classification and some properties traditionally ascribed to God are missing. It must be admitted that the very generality of the classification scheme leaves many important issues unresolved. For example, to say that God knows the world does not specify the nature, the objects, or the mechanics of divine cognition. One finds very different views of omniscience in the works of Aquinas, Molina, and Lequyer. Or again, to say that God includes the world does not settle any disagreements among Ramanuja, Spinoza, Whitehead, Teilhard, and Hartshorne about how world-inclusion or divine embodiment is to be conceived. The subject of God’s inclusion of the world—that is to say, of panentheism—has stimulated much philosophical discussion in the twenty-first century (e.g. Clayton and Peacocke 2004; Cooper 2006).

A topic conspicuous by the scant attention paid to it in Philosophers Speak of God is deism. A deist, like the temporalistic theist, might adhere to ETCK. This way of stating the issue, however, does not bring out the equally important dissimilarities between the doctrines. For example, the temporalistic theism of Lequyer and the deism of Thomas Jefferson are profoundly different in the ways they conceive the world and God’s relation to it. For Lequyer, the very existence of the world makes “a spot (tache) in the absolute which destroys the absolute” (Hartshorne and Reese 2000: 229). Jefferson’s God, on the other hand, is said to be so “far above our power” that nothing we do has an effect on its perfection (Viney 2010: 102).

Another important lacuna in the classification system is the concept of God’s creativity. The expression divine creativity has different meanings, for example, for Plato, Aquinas, Jefferson, Whitehead, and Neville. These delicate issues are simply not addressed in the ETCKW classification. This is not to say that the concept of creativity is not discussed in Philosophers Speak of God, for it is, and extensively so. Moreover, Hartshorne and Reese noted that their classification of doctrines put stress on “consciousness and knowledge but not on volition and power” (2000: 23). They justified this emphasis as a needed corrective to the usual “tendency to put power or causality or eternity uppermost in theological speculation…” (2000: 23–24). We shall see in what follows that Hartshorne explicitly introduces the idea of creatively into his other classification schemes.

Finally, it should be noted that divine love is mentioned only in passing by Hartshorne and Reese. This is surprising in light of Hartshorne’s emphasis in other writings that this is the most important characteristic of God (e.g. 1941: 396, 1997: 70, 167, 213). For Hartshorne, “God is love” was his “ultimate intuitive clue in philosophy” (Hahn 1991: 634 and 700). Early in his career he even said that he strove to show that “all concepts get their meaning from that axiom” (Auxier and Davies 2001: 14). Hartshorne and Reese maintain, in their critical comments on Al-Ghazali, that love is the essence of God (2000: 110) and, in their discussion of Sigmund Freud, they say that the injunction to love one’s neighbor as oneself is a statement of the religious ideal (2000: 482). Thus, divine love functions as something like a regulative ideal in the assessment of theistic doctrines, but it is not an explicit part of the classification system.

Despite its limitations, Hartshorne’s approach to the history of philosophy is conducive to discovering ideas about God that have been marginalized or ignored by the regnant tradition. The classificatory system of Philosophers Speak of God, while being incomplete, allows one a glimpse not only of some historically significant forms of theism but also alternatives not usually discussed. A noteworthy example is Hartshorne’s inclusion of an excerpt from the writings of Lequyer, the first to appear in English (2000: 227–230). Harvey Brimmer, Hartshorne’s student at Emory University, wrote a dissertation on Lequyer. It was, moreover, the mention of Lequyer in Hartshorne’s writings that led to my own work on this much neglected French philosopher (Viney 2009; Lequyer 1998, 1999).

The classification of theistic ideas in Philosophers Speak of God advances discussion of the concept of God by providing a new gestalt that invites a rethinking of issues in philosophical theology. Before Hartshorne and Reese, classical theism held pride of place as the normative theistic doctrine. All other views were considered inferior versions of classical theism. Philosophers Speak of God made a Copernican-like shift by characterizing classical theism (ECK), among others, as a “truncated” version of a more inclusive view, a “neo-classical” understanding of the concept of God (ETCKW). Hartshorne and Reese thereby proposed that neoclassical theism be considered the norm for the concept of God, of which all else is a lesser rival.

The Conceptual Approach

Hartshorne’s conceptual approach to discovering the varieties of theism complements and gives systematic structure to his study of the history of ideas, and holds the promise of remedying some of its shortcomings. From the late 1930s until the publication of The Zero Fallacy in 1997 Hartshorne continually refined the ways in which to think of the logically possible varieties of theism. In the early stages of his thinking Hartshorne focused on the meanings of perfection. As his thought developed he explored the ways in which polar contrasts could apply to both God and the world.

The earliest example of the conceptual approach is the 1940 essay, “Three Ideas About God.” The three ideas are:

• 1. God is in all respects perfect or complete.

• 2. God is perfect and complete in some respects, but not in all.

• 3. God is in no respect entirely perfect.

Hartshorne argues for the merits of the second idea and rejects the other two. What is important for our purposes is that he expresses the second idea in two ways. In the first way he says that “a God both perfect and imperfect will be unchanging in the ways in which he is perfect, and changing in the ways in which he is not perfect” (1953: 160). In the second way, Hartshorne avers that “perfection” has different meanings and it may be incorrect to speak even of a changing God as imperfect. God may be, in some respects, unsurpassable by all others, but in other respects, surpassable, but only by the divine self.

The 1940 essay is the last time that Hartshorne referred to the God in whom he believed as in any way imperfect. (It was a strange oversight when in 1953, in Reality as Social Process, he reprinted this essay without revising it.) In 1941, in Man’s Vision of God and the Logic of Theism, he was much clearer about the meanings of perfection and about “The Formally Possible Doctrines” of God. He says that a God that is unsurpassable by any being, including the divine self, possesses A-perfection, or absolute perfection. A God that is unsurpassable by any being, excluding the divine self possesses R-perfection, or relative perfection. Hartshorne notes that a single being may possess both kinds of perfection provided that it does not have them in the same respect. Thus, rather than saying, as he had a year earlier, that God is perfect in some respects and imperfect in others, he says that God is both A-perfect and R-perfect in different respects.

If one adds the possibility of denying either A-perfection or R-perfection and if one assumes that all aspects of a being must be taken into account, one has an exhaustive classification (Table 2, modified from Hartshorne 1941: 9). If God is in no respect imperfect then (1), (2), and (5) are the theistic options. Ideas of a deity or deities that are merely finite, limited, or even wicked are covered by the other options. The seventh option may also be considered the atheistic alternative insofar as it includes the situation in which there is no being that is in any respect perfect.

Table 2 God as A-perfect or R-perfect

Hartshorne’s classification is an improvement upon most treatments of the theistic question before his time. By introducing the concept of R-perfection he demonstrates that most philosophers and theologians, including those who assume that God can in no way be imperfect, have not considered an important alternative, namely that there could be a perfect form of change in the divine being. Thus, to do justice to the theistic question—including the question whether God exists—one must place beside classical theism (1), views of God that allow for R-perfection as possible ways to conceive the divine reality. Hartshorne himself makes a clean break with classical theism when he refers to God as “the self-surpassing surpasser of all” (1948: 20).²

The categories of Man’s Vision of God are also superior to those later developed in Philosophers Speak of God insofar as they make explicit that perfection is the defining feature of deity. By speaking of perfection rather than eternity, temporality, consciousness, knowledge, or world-inclusion, Hartshorne could give a more prominent role to the idea of divine love. In Philosophers Speak of God he and Reese wrote, “A genuine acceptance of ‘God is love’ is not easily learned, even from Christian—or Jewish—theologians” (2000: 110). The problem is that the dominant theologies in Christian and Jewish circles emphasized God’s A-perfection and ignored the category of R-perfection. God could move the world but was unmoved by it. This involved the difficult if not impossible idea that divine love, unlike human love, is in no way affected by the beloved. By focusing on the logic of perfection (the title of one of his later works) Hartshorne could more easily develop a philosophical theology that takes love seriously as an experience of which God is capable rather than as being, as it was for the classical tradition, a mere behavior of deity. Hartshorne’s form of theism is no mere theological behaviorism. As he would later say, God not only moves others, but is moved by them in the best possible way: God is the most and best moved mover (Viney 2006).

Hartshorne’s next attempt at categorizing theistic doctrines was in “A Mathematical Analysis of Theism” (1943), reprinted a decade later as the epilogue of Philosophers Speak of God. In this article Hartshorne again used the distinction between A-perfection and R-perfection. But now he adds the distinction between God as in some sense independent and creative of the universe and God as inclusive of, and possibly identical to, the universe. This yields a ninefold classification, assuming that both contrasts are represented (AR and CW), and excluding the possibilities where either God or the universe do not exist (1943: 34; 2000: 512) (Table 3).³ This classification preserves the contrast between classical theism (1) and views that allow for R-perfection. The R-perfect options, however, now branch into the three possibilities, (2), (5), and (8). Hartshorne’s position is represented by (5).

Table 3 Hartshorne’s 1943 classification

I noted previously that the classificatory scheme of Philosophers Speak of God omits the concept of divine creativity. The 1943 classification includes this idea but leaves it undefined. The letter C means “God as in some sense independent and creative of the universe” [emphasis mine]. According to this broad interpretation, Plato, Aquinas, Jefferson, Teilhard, Whitehead, and Neville all believe in divine creativity. Thus, where divine creativity is concerned, the 1943 classification only distinguishes doctrines of God that affirm some sense of divine creativity and doctrines that deny it.

Hartshorne’s attempt to think clearly about the logically possible forms of theism began to take its most perspicuous form with the publication of Creative Synthesis and Philosophic Method (1970). He takes his clue from the different ways that classical theism applies metaphysical contrasts to God and the world.⁴ Classical theism (represented in the classifications considered thus far by ECK, A, and A-C) is the view that perfection precludes any principle of potency—God is pure act. By virtue of being pure actuality God is unchangeable in all respects (immutable), has no contingent qualities (necessity), is unqualified by time (eternal), lacks parts (simple) and is nonphysical (immaterial). Another consequence of the classical concept of deity is that God is wholly unaffected by worldly processes (impassible). Finally, according to classical theism, God creates the universe ex nihilo, from no pre-existing material. This creativity is categorically different from any creativity in the creatures.

According to classical theism, God and the world stand on opposite sides of the polar contrasts. For instance, if God is immutable, the world is mutable, if God is necessary, then the world is contingent, and so forth (Table 4). According to Hartshorne, classical theism is monopolar in the sense that it associates God with only one pole of the pairs of contrasts; likewise, the world is characterized by only one side of the lists of contrasts. Hartshorne attributes each pair of contrasts, in different senses, to both God and the world—hence, dipolar theism, one of the names for Hartshorne’s view.

Table 4 Ultimate contrasts: God and the World

The seven contrasts listed in Table 4 are not the only ones Hartshorne discusses. There are also the contrasts absolute/relative, independent/dependent, infinite/finite, cause/effect, object/subject, actual/potential, being/becoming, psychical/physical, and others (1970: 100–101). Formally speaking, each member of a pair is such that it applies to God or it does not. Thus, for each pair there are four possibilities. For example, in the case of the necessity/contingency contrast, either (1) God is wholly necessary, (2) God is wholly contingent, (3) God is necessary and contingent in different respects, or (4) God is neither necessary nor contingent. A similar fourfold analysis applies to the world. The combined possibilities for any pair of contrasts as applied to both God and the world are 16 (Table 5). Each of these 16 possibilities is listed in Creative Synthesis (1970: 266, 271).⁵ Hartshorne says, however, that he did not discover the four-row, four-column arrangement until his ninetieth birthday, with the help of Joseph Pickle at Colorado College (Hartshorne 1997: 42, 84).

Table 5 Necessity and contingency as applied to god and the world

A significant difference between Hartshorne’s presentation of the 16-fold matrix in Creative Synthesis and in his later writings is the interpretation that he gives to the zeros. In Creative Synthesis, the zeros are the atheistic and acosmic positions (1970: 271–72). In later discussions, however, he interprets the zeros more broadly as “God is impossible (or has no modal status)” and the “World is impossible (or has no modal status)” (1992: 18, 1993a: 17, 1993b: 296, 1997: 83). To illustrate the difference between these interpretations consider the position of Willard Quine. He would say that God does not exist, the world does exist, but the world has no modal status. This option cannot be represented as O.n, O.c, or as O.cn since each presupposes modal status for the world. Nor can it be represented as O.o without serious distortion, since Quine does not deny that the world exists. Another illustration of the problem is Robert Neville’s emphasis on apophatic theology. On Neville’s view, the necessary/contingent contrast is a product of God’s creative act; God cannot be characterized as either necessary or contingent, but only as indeterminate, at least prior to the act of creation. Hartshorne’s table, as I present it here, finesses these issues by interpreting the zeros in a strictly formal fashion to mean “neither necessary nor contingent,” leaving open the possibility of further refinement.⁶

Despite the problem in interpreting the zeros, the 16-fold matrix is a substantial advance on Hartshorne’s early attempts at listing the logically possible doctrines of God. First, the table explicitly includes both God and the world whereas his earlier views included the world only implicitly. Second, assuming that some positive statements about God are possible, the 16 positions are exhaustive and mutually exclusive. In other words, one of the alternatives is true and 15 are false. Third, Hartshorne constructs similar tables for other polar contrasts, providing even more detailed distinctions among theistic and atheistic doctrines (1985: 229, 231, 1986: 70, 1993a: 17; cf. Viney 1996: 118). Finally, Hartshorne’s matrices provide a precise method for making distinctions among various types of historically significant world-views: Parmenidean monism or classic Advaita Vedanta  =  N.o; early Buddhist thought  =  O.cn; Aristotle’s theism  =  N.cn; Aquinas’ theism  =  N.c; Stoic or Spinozistic pantheism  =  N.n; LaPlacean atheism  =  O.n; John Stuart Mill’s theism  =  C.n; William James’ theism  =  C.c; Lequyer’s theism  =  NC.c; Russell’s atheism  =  O.c; Hartshorne’s theism  =  NC.cn.

Philosophers who call themselves “free will theists” or “open theists” have occasionally complained that process theologians—and by implication, Hartshorne—are guilty of arguing from a false alternative between classical theism and panentheism, thus ignoring their favored alternative (Nash 1987: 21, 149; Pinnock et al. 1994: 9). Hartshorne’s matrices demonstrate that he is not guilty of this form of reasoning. Moreover, Hartshorne’s matrices provide a method for making distinctions among various types of free will theism. Consider the questions whether God is eternal and/or temporal (E/T) and whether God is immutable and/or mutable (I/M). William Alston, who David Basinger says was probably a free will theist (Basinger 1996: 140), argues that one may admit contingency in God but adhere to the divine immutability and nontemporality (Alston in Cobb and Gamwell 1984: 78–98). Expanding on Hartshorne’s notation we have for Alston’s theism NC/E/I.c/t/m, that is to say, God is necessary and contingent in different respects, wholly eternal and immutable, but the world is wholly contingent, temporal and mutable. Other free will theists, however, accept contingency, change, and time as part of the divine life: symbolically, NC/ET/IM.c/t/m.

Finally, it should be noted that the number of formally possible concepts of God and the world is far greater than Hartshorne seemed to realize. Hartshorne said that “the 16 options become 32 if each is subdivided into those accepting and those not accepting Plato’s mind-body analogy” (1997: 83). While this is correct, the number of formal alternatives leaps to 256 (16  ×  16) if one combines any two pairs of contrasts. More generally, if m equals the number of contrasts one wishes to include in talking about God and the world, then 16^m is the number of formal alternatives available.

I noted in passing the problem in interpreting the zeros in the matrix. Before closing, I would like to note that some headway in addressing this issue might be made by means of what could be called a meta-matrix, specifically, a matrix that addresses the difference between kataphatic and apophatic theology. In keeping, however, with Hartshorne’s method of applying polar contrasts to both God and the world, let us speak not only of positive and negative theology but also of positive and negative cosmology (Table 6). A completely apophatic cosmology, analogous to an apophatic theology, denies that anything positive can be known about the world. Historically, the closest thing to a completely negative cosmology is Pyrrhonian skepticism.

Table 6 Kataphatic and apophatic theologies I

As in Hartshorne’s position matrices, a letter standing alone means “in all respects” and two letters standing together mean “in some respects one way, in other respects another way.” Thus, KA means, in some respects God is known as having positive aspects and in other ways, God is known only negatively. Thus, column I and row 1 represent a completely positive theology or cosmology respectively; column II and row 2 represent a completely negative theology or cosmology respectively; column III and row 3 represent a pairing of positive and negative theology or cosmology respectively; the zeros can now be understood straightforwardly as, in column IV and row 4, the atheistic and acosmic options respectively.

This matrix does not bring out the primacy that a particular philosopher might give to K or A. Hartshorne, for example, believes that in theology, primacy should be given to positive ascriptions; Maimonides seems to favor the negative. One could solve this problem by this convention: Let KA  =  primacy given to kataphatic theology; let AK  =  primacy given to apophatic theology (similarly for the cosmological cases). This results in an additional column and an additional row, yielding a 25-fold matrix (Table 7).

Table 7 Kataphatic and apophatic theologies II

This matrix differs from the previous one in the third and fourth columns and rows, where emphasis on the positive or the negative has been stressed. Hartshorne’s view is expressed by KA.ka whereas Neville’s view is somewhere in the fourth, or AK, column. The meta-matrix is not nuanced enough to capture the differences among philosophers about literal versus analogical ascriptions, or between anthropomorphic and non-anthropomorphic language. As in the case of concepts of God, however, where different matricies express different polar contrasts, so in the case of theological language, more than one matrix can be used to map the formal space of different dimensions of the problem. It would not be difficult, following Hartshorne’s method, to construct separate matricies for the formally possible combinations of analogical and literal language, or anthropomorphic and non-anthropomorphic language, as they apply to God and the world.

Conclusions

We have seen that Hartshorne’s early attempts at classification omitted or did not do justice to the idea of divine creativity. His last classification scheme allows for modest improvement. Since Hartshorne includes contrasts as they apply to the world, one may speak of the world—or individuals within the world—as being creative. Thus, classical theism, which says God is creator and worldly individuals are created, can be contrasted with forms of theism that allow creativity in the creatures. For instance, Lequyer speaks of “God, who created me creator of myself” (1952: 70). The idea that genuine creativity is found in both God and non-divine beings is easily symbolized (C.c, where “C/c” represent creativity) in Hartshorne’s final classification system.

It is less clear whether Hartshorne’s schema can adequately address the question of deism. Deism, unlike other forms of theism, affirms divine creativity but denies God’s continued involvement with the world. One might argue that deism, by denying R-perfection but affirming a certain independence of the world from God’s activity, combines the worst of classical theism and the best of Hartshorne’s neoclassical theism. On the other hand, deism, like classical theism but unlike neoclassical theism, is deterministic. The deistic doctrine of the world’s independence from God is tantamount to a denial of miraculous intervention from the realm of the supernatural. Hartshorne too denies miraculous intervention, but this is because the distinction between the natural and the supernatural is alien to his neoclassical metaphysics.

It is too much to ask of Hartshorne’s matrices to solve differences of doctrine in philosophical theology. After all, the matrices, as Hartshorne presents them, are arrays of formally possible alternatives and are limited to issues in philosophical theology. They touch, only by implication, broader issues in the philosophy of religion which, following Robert Neville, I define as “the critic of abstractions in religions” (Neville 1995: 169). Religions come in many varieties, not all of them theistic. As Neville points out, it is parochial to assume that, for religions that deny or downplay the existence of God, their non-theistic perspective is the most important fact about them (Ibid.,: 173).⁷

Their limitations notwithstanding, Hartshorne’s matrices represent a significant advance in philosophic understanding. It is useful, in the first place, to be able to map alternatives in philosophical debate if for no other reason than to avoid the fallacy of false alternatives. The matrices, used in concert with the historical approach, can illuminate the theological landscape by spotlighting not only well-known doctrines but also lesser known options. What Hartshorne calls the monopolar prejudice of classical theism (cf. Table 4) is the familiar and dominant way of thinking of God. A genuinely dipolar theism is unlikely to get a fair hearing until it is seen as a legitimate option. The matrices help to bring the dipolar options into clear relief. Indeed, we have noted that the matricies relativize classical theism by showing that it can be construed as a truncated version of neoclassical theism.

Equally, the matrices are useful in helping one see what is being denied by any particular theological construct. Any alternative accepted is 15 (or 24, as in our last matrix) alternatives rejected. According to Hartshorne, many philosophical errors are hidden in what philosophers deny (cf. 1997: 166). At the very least, a philosophical position can be as interesting for what it denies as for what it affirms. Hartshorne’s matrices can serve theistic and non-theistic religions by outlining their denials. One may recommend Hartshorne’s method as a necessary prolegomenon for dialogue among competing theological perspectives.

If Hartshorne’s method, suitably refined, allows one to chart most historically significant concepts of God, it is no less effective in providing a map of some unexplored areas of philosophical theology. For example, Hartshorne’s final classification scheme explicitly includes the world and raises the question how metaphysical contrasts apply to it. It is often as important to reflect on the relevance of one’s doctrine of the world for one’s idea of God as it is to reflect on the relevance of one’s doctrine of God for one’s idea of the world. Aristotle understood this when he denied that an immutable deity could know a mutable world. Aquinas too understood this when he inverted the usual dependence of the knower upon the known to show how his immutable God could know a mutable world (see Hartshorne’s discussion of this in 1976). We have also seen that position matricies can be extended to include issues in the philosophy of language as this applies to theological discourse.

Hartshorne’s matricies highlight the fact that not all forms of theism require supernaturalism. Traditional theism posited a two-tiered universe of the creator (God) and the created order (the creatures). The basis of the distinction is that God alone has the power to create, to bring something from nothing, or from no pre-existing material. In Hartshorne’s theism—as well as some others—all creativity, including God’s, requires antecedent “material” from which to create. This does not mean that principled distinctions cannot be made between God and the creatures, but the lines need not be drawn on the concept of creativity itself. Hartshorne argues that the differences lie elsewhere: for example, in the scope of creativity—God affecting and being affected by all others; the creatures affecting and being affected by some but not all others; and in the quality of creativity—God’s responses are ideal and creaturely responses are always less than ideal (see especially Hartshorne 1967: chapter 2).

The issue of unexplored alternatives brings us full circle to Lovejoy’s insight about the relationship between the questions whether God exists, what we mean by “God,” and even what we mean by the language that we use about God. Hartshorne’s matrices are a dramatic demonstration that the question “Does God exist?” is at best vague (if one hasn’t a clear vision of the alternatives) and at worst loaded (if one tacitly privileges one concept of God or one mode of theological discourse). If there are many forms of theism, it follows that there are many forms of atheism—as many as there are forms of theism to deny. The significance of this insight for the life of faith cannot be underestimated. Hartshorne’s “history” of God frees the imagination to think of the divine in ways often bypassed by the dominant systems of belief. Theology may then prove to be as dynamic as the God in whom Charles Hartshorne believed.