(DOC) Goris Divine Foreknowledge edited

1 Divine foreknowledge, providence, predestination, and human freedom Harm Goris Aquinas maintains that divine foreknowledge, providence, and predestination are compatible with creaturely contingency and freedom.1 He does so throughout his career and in many of his texts, from the early commentary on Peter the Lombard’s Sentences, written in the mid-1250s, up to his very last writing, a letter to the abbot of Monte Cassino, written toward mid- February 1274 about three weeks before he died. 2 It is, however, less clear (1) how exactly Aquinas argues for that compatibility, and (2) whether his arguments are cogent. The first question, how to interpret Aquinas’ texts, was one of the points over which a fierce debate arose among Catholic theologians around the end of the sixteenth century, known as the De Auxiliis controversy. The theological issue at stake was the coherence of God’s foreknowledge, predestination, the efficacy of grace, and human freedom. In this debate, each of the two rival parties claimed that its position was supported by Aquinas, who had been proclaimed Doctor of the Church by Pope Pius V in 1567. But both parties also assumed that Thomas had not fully explicated his view, and they sought to supplement the Thomistic doctrine. The Dominican party, led by Domingo Bañez, introduced the notion of “physical premotion” (praemotio physica). According to their theory, God has predetermined in the eternal decrees of his will to concur in an irresistibly efficacious way to the activities of creatures in time, even when they act freely. Their major adversaries were the Jesuits Luis de Molina and Francesco Suárez, who saw in the “physical premotion” a denial of human freedom. As an alternative, they developed the theory of the scientia media, the contingent knowledge God has of what a possible creature would actually choose, given any possible set of circumstances. The controversy came officially to an end in 1607, when the Pope intervened. Both parties were summoned not to charge each other with heresy – either Calvinism or (semi-)Pelagianism – and to submit all writings on the subject to the Inquisition prior to publication. The Holy See would promulgate its decision in the matter at an opportune moment – which, apparently, is still pending. 3 Although officially the dispute was left undecided, the Dominican Bañezians emerged victorious, in the sense that their theory became generally recognized as the Thomistic doctrine. From the 1960s onward, analytical philosophers of religion regained interest in the classical question if and how God’s foreknowledge and providence are compatible with human freedom. While the Molinist theory of middle knowledge features prominently in this present-day debate, Thomas’ views have received relatively little attention. 4 It seems that the predominance of the Bañezian interpretation, with its deterministic implications in particular, is the major reason for this. Only in recent times some have begun to question the identification of the Bañezian theory with Thomas’ own view. Apart from the systematic problem of determinism, it has been pointed out that the expression praemotio (or: praedeterminatio) physica does not appear in Aquinas’ texts at all. Moreover, if God’s will and causality suffice to account for the certainty of God’s foreknowledge, then the argument that Thomas usually offers, viz. God foreknows because all of time is present to his eternity, becomes superfluous. 5 Finally, it has been argued that Bañez’ theory does not do justice to God’s transcendence because it suggests that God is merely the first element of a physical chain of movement.6 In order to get a better grasp of Thomas’ own views, I shall first present some elements from his discussions on theological language. The analysis of the terms “foreknowledge,” 2 “providence,” and “predestination” will make it clear that Thomas distinguishes two different problems when it comes to the compatibility of divine agency with contingency in the world. One has to do with the diachronic relation of foreknowledge and future contingents, and the other with the synchronic relation of a necessary cause and its effect. How Aquinas deals with each of these problems will be discussed in the subsequent sections. 1. Divine names Thomas does not aim at construing a watertight, logically deducible theological system from scratch. He analyzes and reflects on what has been given to him: the Catholic faith and its religious language, as it is used and handed down by Scripture and the Church. As a theology professor, he wants to gain, and pass on, insight into the mysteries of faith, show the non- contradictory coherence of the nexus mysteriorum, and explore the limits and possibilities of our understanding in divinis. In this life, we cannot understand what God is (quid Deus sit), and, hence, all the words we use for God (“divine names”) are radically imperfect. No one concept or name captures and represents adequately the mystery of God. In Thomas’ view, all divine names refer to the one, simple divine reality, but this does not mean that they are synonymous. For the conceptual content that each name expresses differs, giving us only a partial, limited and opaque inkling of the mystery of divine perfection. Each name leaves God’s being, or wisdom, or knowledge as “not comprehended and exceeding the signification of the name.” 7 This basic insight permeates the whole of Thomas’ theological enterprise. He is always very aware of the limits and radical imperfection of our language and understanding in divinis, but he also insists that this does not prevent us from speaking both properly and truly of God.8 In analyzing the semantics of divine names, Thomas uses several ways for classification. One of them is the distinction between (relative) names that signify immanent and those that signify transient acts. Logically, immanent acts do not require at the same time the real existence of the objects; a person may be said to know or will things that exist only intentionally. On the other hand, the grammar of transient acts like “causing,” “creating” or “justifying” does imply the real existence of the external effect. When used in divinis, this distinction coincides with divine names that are predicated of God “from eternity” (ab aeterno) and those that apply to him in time (ex tempore).9 “Creating” and “causing” belong to the latter. Furthermore, Thomas points out not to think of God’s causing as the setting in motion of a chronologically successive series of events that, after some time, results in a final effect. We should speak of God’s causation as instantaneous: at the very moment God causes something, it is.10 This explains why we can say that God knows from eternity what will happen tomorrow, but not that he creates from eternity tomorrow’s events. Thomas would consider “eternal causation” to be a misnomer. Again, this does not imply any real distinction or change in God. God’s acts of knowing and willing do not differ in reality from his act of being, nor does he create or govern by any supervenient act(s). 11 The distinctions are only conceptual. They come from the inability of our language and understanding to capture fully God’s perfect being and activity. Saying that God “knows” or “wills,” does not yet express that God communicates being to the object known or willed so that it exists in its own right. For that we need other verbs like “to create” or “to cause.” Strictly speaking, “willing” belongs to the names that signify immanent acts and that are said of God from eternity. But Thomas acknowledges that, in practice, theological language is not that rigid. He states that, unlike the term “knowing,” “willing” has also the connotation of 3 causality.12 For it expresses that God’s acting ad extra is free, is directed to the good, and does not imply any real change in God. Therefore, it can be used for signifying God’s efficient causation of creatures in time.13 In a discussion on God’s will Thomas distinguishes between the intentional and the external object of the will: The will has a twofold relation to what is willed. First, insofar as it is willed; secondly, insofar as it is to be actually produced by the will. And the latter presupposes the first relation. For we understand first that the will wills something; next, from the fact that it wills it, we understand that it produces the thing in reality, if the will is efficacious.14 Thomas continues by pointing out that, when considered as (intentionally) willed, the object of God’s will is not necessary. For God is free to choose. But when God’s will is regarded as the cause of the object, then the object exists necessarily, for nothing can thwart divine causation: For the first relation of divine will to what is willed is not necessary in an absolute sense, because of the lack of proportion of what is willed to its end … But the second relation is necessary because of the efficacy of divine will; and therefore it follows necessarily, if God wills something by the will of pleasure, that it happens.15 It is the second relation which is relevant to the question of determinism: God’s will as cause ad extra of created reality, and, hence, signified as a transient act in time. The terms “providence” and “predestination” express God’s plan to lead creatures to their end. In the case of “providence,” it is the ordering of all beings to their end in general. “Predestination,” on the other hand, is a particular part of providence. It regards specifically the ordering of rational creatures to a supernatural end, which they cannot reach by their own, natural powers: the beatific vision of God himself. Because “providence” and “predestination” signify intellectual planning and intention, they belong to both God’s knowledge and his will. 16 Usually Aquinas classifies “providence” and “predestination” as immanent acts from eternity, and distinguishes them from their transient counterparts in time. “Governance” (gubernatio) signifies the execution of providence, while “justification by grace” and “glorification” signify the effects of predestination in time.17 But, as we shall see below, when it comes to the problem of determinism, Thomas also takes into account the determinism that seems to follow from the causal aspect of God’s will, that is, from “providence” and “predestination” taken as transient acts and as causes of what happens in time. Another important element in mapping out the logic of the divine names “foreknowledge,” “providence” and “predestination” concerns the prefix “prae.” Augustine, Boethius and Anselm had argued that, because God is timeless, he does not really have fore- knowledge or fore-sight (praevidentia). Aquinas agrees that God is outside of time, but in his view this doesn’t render the use of “prae” false or improper. For it is we, temporal beings, who speak about God, and the things that we say God foreknows or predetermines, are future for us. 18 That is why the terms “foreknowledge,” “providence” and “predestination” can be used properly, and only with regard to the future.19 In the question how foreknowledge, providence and predestination can have contingent things or events as their objects, Aquinas distinguishes two problems. One has to do with the relation of knowledge to future contingencies, and the other with the relation of a necessary cause 4 to contingent effects. When Thomas discusses the problem of foreknowledge for the first time in the commentary on the Sentences, he writes: For contingencies seem to elude divine knowledge for two reasons: Firstly, because of the order of cause to what is caused. For the effect of a necessary and immutable cause seems to be necessary; therefore, as God’s knowledge is the cause of things and as it is immutable, it does not seem that it can be of contingencies. Secondly, because of the order of knowledge to what is known; for, as knowledge is certain cognition, it requires from the notion of certainty, even if causality is excluded [italics by me, H.G.], certainty and determination in what is known, and contingency excludes this; that knowledge, because of its certainty, requires a determination in what is known, is clear in our knowledge, which is not the cause of things, and in God’s knowledge with regard to evil things.20 The reference at the end of this quotation to human knowledge and to God’s non-causal knowledge of evil specifies again that the concept of “knowledge” by itself does not include causality. According to Thomas, it poses its own kind of problem. This is why he usually separates the discussions of the two problems. The problem of causality is dealt with under the heading of God’s will, and the problem of knowledge is discussed within the context of divine knowledge. As the notions of “providence” and “predestination” imply both God’s knowledge and causal will, they combine the two problems.21 What exactly are the two problems? The text from the commentary on the Sentences, quoted above, states that the notions of “knowledge” (or “certain, non-conjectural cognition”) and of “necessary cause” require each a determination or certainty of their objects which is excluded by the notion of “contingency.” In another text, while discussing the certainty of predestination, Aquinas distinguishes the two kinds of determination as follows: There is a twofold certainty: one of cognition, and the other of order. There is certainty of cognition when cognition does not deviate in any way from what is found in reality, but judges in such a way about reality as it is. Because a certain judgement is held about things especially through their cause, the noun ‘certainty’ has been transferred to the order of a cause to its effect. Consequently, the order of the cause to its effect is said to be certain when the cause produces the effect infallibly.22 Elsewhere Thomas explicates that it is not the contingent as such, but the contingent insofar as it is future and not present or past, that lacks the determination or certainty which knowledge requires.23 An example may clarify the distinction between the two problems. Let us take “I shall have waffles for breakfast tomorrow” as a neutral phrase that can be the object of different acts. First, consider it as an object of knowledge. What I am having for breakfast in general is contingent; it depends on my choice. But when it comes to particular events, there is a difference between, on the one hand, past and present and, on the other, the future. My choice for yesterday’s breakfast was contingent, but precisely because it is now past, it is determinate and is expressed by the true statement “I had waffles for breakfast yesterday.” Therefore, it is very well possible for someone to know that I had waffles for breakfast yesterday. This is not the case with my choice for waffles tomorrow. That event is not determinate. I may be inclined now to take a 5 cheese omelette tomorrow, but in between now and tomorrow morning I may change my mind (or I may even die). What (or even, if) I shall choose tomorrow is not determinate. Consequently, the statement “I shall have waffles for breakfast tomorrow” cannot be said to be true, nor to be false. Therefore, it cannot be the object of knowledge, for knowledge is, by definition, of what is true. Next, consider the same phrase as an object of will. Then it need not be determinate and true. If, however, it is stipulated that the will is an infallibly efficacious and instantaneous cause, and if “I shall have waffles for breakfast tomorrow” is not considered as an intentional object, but as a real, external event, then my having waffles tomorrow will be caused tomorrow in such a way that it will not depend upon my choice anymore and I shall not be able to refrain from it. Necessity is an ambiguous term. We speak, for example, of logical, hypothetical, moral, natural, metaphysical and causal necessity. It is important to note that the two problems at stake involve two different kinds of necessity. Divine causation seems to imply causal necessity, but the problem of foreknowledge concerns temporal or factual necessity. It is a kind of necessity that has to with the nature and passage of time. Intuitively, we think that past and present as such have some kind of necessity and determinateness that the future has not. 24 This is what in present- day philosophy of time is spelled out in the so-called “tensed” or “dynamic” or “A-series” view on time with an “open” future. In general, a tensed view considers the monadic and changing notions of “future,” “present,” and “past” as the basic, metaphysical characteristics of temporal events themselves: an event that was future, is now present, and will be past. It has to be noted, however, that on a specific version of the tensed theory, viz. one with an open future, a future event cannot be identified as an individual, particular event in the same way as present and past events can be. One may only refer to types of future events. We can imagine this version of the tensed theory as the tip of a pen – representing the present – which constantly moves on and draws the line of the past behind it. In front of the tip there is nothing: it is blank, the future is open.25 The rival theory, called the “tenseless” or “static” or “B-series” view, reduces the dynamic, monadic categories of “future,” “present,” and “past,” to the unchanging relations of “later than,” “simultaneous with,” and “earlier than.” For example, instead of “Caesar was murdered in the past” the tenseless view holds “Caesar is murdered 44 years earlier than Christ is born,” where the two finite verbs “is” are logically understood as tenseless, though they have grammatically a present tense.26 The tenseless theory considers time as a fourth dimension, and states that what is earlier or later relative to this moment, is in itself as determinate and factual as what happens now. It pictures time as a static, straight line that includes also all that is future relatively to this moment.27 On such a theory, one may admit that not all events, whether past, present or future relative to this moment, are either causally or logically necessary, and that some of them could have been counterfactually otherwise.28 But on the tensed view with an open future, it does not make sense to talk about the future “that could counterfactually have been otherwise.” For on this view there are no future facts, and, hence, no future counter-facts: the future lacks the factual or temporal necessity and determinateness of the past and present. By carefully mapping out the semantics of divine names, Thomas concludes that there are two distinct problems regarding the compatibility of foreknowledge, providence, and predestination with creaturely contingency and freedom. The first one concerns the diachronic relation between foreknowledge and a future contingent object, and the other one the synchronic relation between a necessary, instantaneous cause and its contingent effect. For reasons of clarity and convenience, I label them “temporal fatalism” and “causal determinism.” How does Thomas deal with each problem and are his arguments convincing? 6 2. Temporal fatalism At first sight, Thomas’ answer to the question how God can know future contingents looks quite simple. He adopts Boethius’ well-known solution of God’s “eternal vision” and the presence of all things to eternity. God, Boethius argues, is eternal, outside of time, and the whole stretch of time, past, present and future, is equally present to him and lays open to his gaze. He illustrates this with a spatial metaphor: God is like someone who observes from on high a procession that moves along down on the plain below. A person who is part of the procession, can only see those that have preceded him, but not the ones that will come after him. But from his elevated position, the observer oversees in one glance the whole procession.29 Three important objections have been raised against this solution: it is logically incoherent, it presupposes metaphysically a tenseless view of time, and it makes God’s knowledge observational and dependent, which is theologically unacceptable. Anthony Kenny is famous for the logical reduction to absurdity of Boethius’ solution. “Present to” or “simultaneous with,” he points out, is a transitive notion: if A is present to B, and B is present to C, then A is also present to C. Consequently, if all of time is present to God, then every temporal event must also be present to every other temporal event. 30 In short, the whole of temporal extension would collapse into one singularity. Eleonore Stump and Norman Kretzmann tried to overcome Kenny’s objection by introducing the notion of E(ternity)-T(ime)-simultaneity in a pivotal article published in 1981. 31 It provoked a lively debate, in the course of which the notion was refined and redefined. 32 It is beyond the scope of this paper to give a detailed analysis of the whole debate, but the gist of ET- simultaneity is as follows. Kenny’s argument is based on ordinary language, in which we talk about “simultaneous with” and “present to” as relations that hold between temporal events. Stump and Kretzmann draw attention to the fact that in Boethius’ solution one of the relata is not temporal but eternal. Therefore, we have to take into account two distinct and irreducible reference frames or modes of existence, time and eternity. According to Stump and Kretzmann, an event in time is ET-simultaneous with the eternal God, without becoming eternal itself (E- simultaneous) nor simultaneous with other temporal event (T-simultaneous). In other words, because it presupposes two different reference frames, ET-simultaneity is not transitive: if temporal event x is ET-simultaneous with the eternal God and if temporal event y is also ET- simultaneous with the eternal God, it does not follow that x and y, both of which are in one and the same reference frame, are ET-simultaneous (or T-simultaneous) with each other. Just like Boethius’ metaphor of the observer who watches from on high, does not exclude the spatial extension and succession of the procession in the plain, in the same way ET-simultaneity allows for temporal extension and succession. How this works will become clearer if we take into account the metaphysical presupposition of ET-simultaneity. Although Stump and Kretzmann claimed that their theory is compatible with a tensed theory of time, Craig has argued convincingly that it isn’t. 33 It presupposes a tenseless view on which all events, whether they are (tenselessly) earlier than, simultaneous with or later than this moment, are (tenselessly) in themselves ontologically on a par and equally real. This does not mean that they are temporally on a par and equally present in themselves. If that were the case, Kenny’s absurdity would follow. On the tenseless view, each event has its own temporal location and extension within the tenselessly existing time-space matrix of the “block-universe.” Moreover, on the tenseless theory of time, it does not make sense at all to speak of an event being “present in itself”: an event is present to, or better simultaneous with, another event, but not present in itself. So it seems that the only way to save the Boethian solution is to adopt a tenseless 7 view on time. God, then, is conceived as being outside of, and encompassing, the whole time- space matrix, in the same way as the present moment encompasses all of space. 34 In order to prevent the misunderstanding that leads to Kenny’s absurdity, one might better rephrase Boethius’ statement that all events in time are “equally present to God” into “all events are (tenselessly) equally real to God.” This might also be the reason why Thomas sometimes qualifies the presence of temporal beings and events to God by inserting terms like “as if” (quasi) and “in a certain way” (quodammodo).35 Apart from the fact that it seems the only way to account for God’s foreknowledge, there is another reason that makes a tenseless theory of time attractive for theologians: it can account for God’s knowledge of what is future to us without suggesting that he derives it from observation. For, if the whole time-space continuum exists tenselessly, it makes also sense to say that it is caused tenselessly. Moreover, the tenseless view seems to offer an easy way for defending divine immutability. If temporal becoming is illusory, we can imagine the whole of reality having come forth “at once” from an immutable cause. In accordance with Christian doctrine and most present-day scientific cosmologies, its temporal extension can very well be limited, having a begin- and an end-term, but it would exist as tenselessly and immutably as its temporally non-extended cause. In recent years, a number of scholars have claimed that in fact Boethius and Aquinas adopted a tenseless theory of time, though they were not aware of it. 36 Consequently, there would be a serious flaw in the way they proceed. For both raise the problem of foreknowledge by presupposing a tensed theory with an open, indeterminate future, but allegedly solve it by adopting a tenseless theory, in which whatever is later than this moment, is in itself as determinate and real as what is earlier than this moment. In the end, they would accept what they intended to refute, viz. temporal fatalism. Moreover, a tenseless theory is not consistent with their view on statements (propositions) about future contingents. In commenting on the famous chapter nine of Aristotle’s De Interpretatione, Thomas adopts the same interpretation as Boethius. Statements about future events that are already fully causally determined by present-day tendencies and laws of nature, are “determinately true” (determinate verum), but statements about future contingents are not. There has been much discussion on how to read the expression “determinately true.” The major reason for the confusion among present-day commentators seems to be that Aristotle and the Scholastics base their views on so-called indexical and temporally indefinite statements, whose truth value depends on when, where and by whom the statement is uttered. Most modern logicians convert such statements (e.g. “I was here yesterday”) into a non-indexical, temporally definite one (e.g. “Harm Goris is in Utrecht on July 24, 2001”). It can be argued, however, that Thomas and other Scholastics do take temporal definiteness into account, while, at the same time, considering indexical tenses essential to statements.37 “Not determinately true,” then, means that statements about future contingents do not have a truth value: they are not true, but they are not false either. They have the disjunctive property of being either-true-or-false, but this property is not truth-functional and the truth values are not distributed. 38 The lack of a truth value is not because of our ignorance, as Boethius explicitly points out, but stems from the “variable and indeterminate outcome of the things themselves.”39 This metaphysical indeterminateness of the future is precisely what the tenseless theory of time denies. Despite the apparent systematic advantages of a tenseless view on time, I think that it is possible to give a coherent interpretation of Thomas’ texts that (1) confirms God’s foreknowledge, (2) accepts a tensed theory of time with an open future and, hence, denies 8 temporal fatalism, (3) does not make God’s knowledge dependent on observation, and (4) coheres with Thomas’ basic theological insight that, in this life, we cannot conceptualize God. As we saw earlier, Thomas sometimes says that all of time is “quasi present” to God. This may be understood as meaning “tenselessly real” to God. But it is possible to give a different interpretation, one that is not based on a tenseless mode of being of creatures, but on the unknowable mode of God’s being: eternity. 3. Eternity Thomas discusses eternity in the context of his “negative theology”: it indicates how and what God is not.40 Aquinas does adopt Boethius’ definition of eternity as the “complete possession all at once of illimitable life,” but he reinterprets it in a negative way. Eternity means that God’s being has no beginning, no end, and no succession. 41 From Boethius, Thomas also borrows the Neoplatonic image of eternity as the “standing now” (nunc stans). Stump and Kretzmann criticized the idea of eternity as a nunc stans. Eternity, they say, is not “a frozen, isolated and static instant,” but had better be conceived as “atemporal duration.” 42 Like their suggestion of ET-simultaneity, the notion of “atemporal duration” also led to much discussion, the main point of criticism being that it is not coherent: duration, that is, extension, without any distinguishable parts (because atemporal) is no extension at all.43 Thomas is somewhat ambiguous about predicating “duration” of God. On the one hand he says that it implies extension (distensio), and should, therefore, not be said of God. On the other hand, he states that eternity does signify duration because of God’s actuality. 44 Instead of concluding that Thomas contradicts himself, I prefer to read his words analogously within the context of a negative theology. In commenting on a text from the Neoplatonic Book on Causes (Liber de Causis) that “there are two kinds (species) of durability, one eternal and the other temporal,” Thomas distinguishes the two as follows: These perpetual durations differ in three ways. Firstly, eternal perpetuity is fixed, standing immobile, while temporal perpetuity is flowing and mobile … Secondly, eternal perpetuity is all at once, as it were gathered in one, but temporal perpetuity has successive extension according to earlier and later, which belong to the notion of time. Thirdly, eternal perpetuity is simple, existing as complete by itself, while the whole or completeness of the temporal is by different, each other succeeding, parts.45 Thomas gives here an interpretation of eternal duration that is very similar to, if not identical with, Stump and Kretzmann’s atemporal duration. But he makes such strong differences between temporal and eternal duration that “duration” in fact can no longer be seen as a univocal term, nor temporal and eternal duration as two species of a generic duration. “Duration” becomes an analogical term. In reaction to some criticisms, Stump and Kretzmann too acknowledged the analogical character of “duration” in atemporal duration.46 But if this is so, why not admit the same about “now” in the classical expression “standing now”? Then the standing now is no longer a static, frozen, isolated instant, just as atemporal duration is neither a tenseless extension nor a tensed succession. In other words, if taken analogously, the “standing now” and “atemporal duration” are two sides of the same coin. God endures, we may say, but not by temporal extension or succession, and he is present, but his presence is not limited by past and future. The 9 words “standing” in “standing now” and “duration” in “atemporal duration” remind us not to think of God’s existence as a static, frozen, isolated instant, nor as an evanescent instant which elapses as soon as it occurs, nor as the atemporality of abstract entities like numbers or universals. On the other hand, the words “now” and “atemporal” in the two formulas tell us not to conceptualize God’s existence as tenseless extension or as tensed succession. This negative/analogous interpretation of eternity is mirrored in our talking about the presence of temporal beings to God. The relative expression “being present to…” has an analogous meaning when God fills in the blank. We can understand that a temporal being is present to another temporal being or event, or that it is past or future to it. But we don’t know exactly what it means when a temporal being is said to be present to the eternal God. For if we did, we would also understand what God’s own mode of being, eternity, is. And this, Thomas repeats over and again, we do not know, at least not in this life. It doesn’t follow that “being present to God” is just idle and nugatory talk. We may truly say that a temporal event is now present to God, but not that it ever was future to God or past to God. As Thomas says “nothing can coexist by way of presence with the eternal, unless with the whole of it; for the eternal does not have successive duration.” 47 It is not the case that “something future” only means that it is (tenselessly) later relative to this moment; whatever it will turn out to be, it is now future and indeterminate in itself. But we cannot say that it is also now future to God. Related to God, it is present and determinate, not because of its own, alleged tenseless existence, but because of God’s transcendent, eternal mode of being. Even if divine foreknowledge is compatible with an open, indeterminate future on the basis of God’s eternity, the question remains whether this does not imply that God’s knowledge is causally dependent on their presence. Is it the case that the presence of temporal beings and events to God’s eternity, is a necessary condition for his observation of these beings, as Boethius’ visual metaphor suggest? Although he doesn’t address this question in particular, Thomas would reject any suggestion of God’s knowledge being dependent. Some have argued that what Thomas means by “the presence of all of time to God” is not the real (sometimes called “existential” or “physical”) presence of temporal beings themselves, but their intentional (“epistemological” or “objective”) presence through the divine Ideas, insofar as determined by the divine will. Bonaventure and other Franciscans read Boethius in this way, but most commentators agree that Thomas means the real presence of all temporal beings to God and that this presence justifies the epistemological claim that God’s knowledge is related to its objects as present objects.48 For Thomas, God’s knowledge of temporal beings is indeed practical and creative, but this causality is not expressed by the notion of “knowledge” as such. “Knowledge” is signified as an eternal, immanent act while “causation” is signified by a transient act in time. The point is that one has to distinguish between the question what the source is of God’s knowledge of temporal reality, and the question what conditions that reality must meet in order to qualify as object of God’s knowledge. Future contingents as such lack the determinate being that is required for knowledge. God cannot know them as such, that is, as future, not because of a deficiency on his side, but on their side.49 But, as he exists eternally, whatever is future both in itself and to us, is not future, but present (in an analogical sense) to him. Consequently, God can know future contingents, but, as Thomas puts it, “not under the concept of future” (non sub ratione futuri). This does not mean that he does not know that they are future. For then his knowledge would be either false or incomplete. The expression “not under the concept of future” concerns the mode of knowing, not the mode of being of the external objects. The human intellect is subject to time, and that is why we know temporal things in a temporal mode, viz. under the notions of past, present and future: our intellect knows by forming tensed propositions. But according to Thomas, 10 the isomorphism between the mode of being of the external objects and the mode of knowing is not necessary. Objects, he often states, are known according to the mode of the knower, not according to the mode of their being.50 The human intellect, for example, knows material, concrete things in an immaterial, universal way. Likewise, Thomas states, the divine intellect knows temporal beings in his own, atemporal, eternal way, which we cannot understand or express.51 Therefore, if we say “God knows that future contingent x will be,” we signify God’s knowledge inevitably in a human way, by using the tensed proposition “x will be.”52 It doesn’t follow, however, that x exists tenselessly and determinately, nor that the proposition by itself is true. Such consequences are only drawn if we are not aware enough of the irremediable imperfection of our understanding and speaking of God, and if we forget our inability to conceptualize and express his transcendent, eternal mode of being and knowing. 4. Causal determinism If God’s will is the all-encompassing cause, which brings about everything and cannot be hindered, is then all that is, or happens, necessary in the sense of causally determined? As we saw earlier, this problem arises when the divine will is considered as the efficient cause of what happens in time. One might also focus on God’s will as free, with the apparent consequence that everything is contingent. Occasionally, Thomas addresses this problem: Although everything depends on God’s will as on the first cause, which is not necessitated in its operation unless on the supposition of its intention, still absolute necessity is not excluded from things because of this, so that we would have to acknowledge that all things are contingent – what someone might think from the fact that things have not come forth from their cause by absolute necessity.53 But more often Thomas directs his attention to the irresistible efficacy of God’s will, and to its compatibility with created contingent causes in particular. Following Aristotle, Thomas distinguishes two kinds of created contingent causes: voluntary causes and natural causes that occasionally do not produce the effect they are determined to by nature. The latter may occur for different reasons: another cause intervenes, the agent cause has some defect, or the matter upon which the cause acts is not suitably disposed. 54 Both kinds of contingent causation have their own specific problems, with which Thomas deals in separate discussions. He speaks about human will in his anthropological texts, moral treatises, and the doctrine of grace and justification. Coincidence in nature is treated in Thomas’ works on natural philosophy and on the theory of science. But in discussing explicitly the question how God’s causation can go together with contingent causation by creatures, he always lumps both kinds of contingent causation together, and aims at offering a general, overall framework for refuting the apparent incompatibility and for indicating the transcendence of the divine cause. It seems that Thomas only gradually developed his view on the transcendence of divine causation.55 In the commentary on the Sentences, he uses Aristotle’s model of ordered causes: the first, remote cause may be necessary, in the sense that it always operates in the same way, but if it produces its effect by mediation of a secondary, proximate cause that is contingent, then the effect will also be contingent, for an effect always takes the condition of its proximate cause. Thomas takes as an example the necessary movement of the sun, which causes a tree to bloom. Yet, he says, the blooming of the tree is a contingent event, for its proximate cause is the 11 vegetative force of the tree, which is a contingent cause and may fail. But Thomas realizes that the comparison with divine causality breaks down: though the movement of the sun cannot be hindered, it can go together with a defect of the secondary, proximate cause, so that the effect does not take place. This, Thomas admits, cannot be the case with God’s causation. The unsatisfactory solution Thomas comes up with is that the ultimate effect is still certain, not because of God’s causation, but only because of his eternal knowledge, to which all is present.56 The next stage in Thomas’ development is that he attributes a kind of statistic certainty to God’s causality. It is found in the discussion on the certainty of predestination in the De Veritate, question six. He says that if God’s causal influence is mediated through a created, necessary cause, then the ultimate effect will follow in each particular instance. But when the created, proximate cause is a contingent one that occasionally fails, the ultimate effect only comes about by and large (in universali). This explanation, Thomas says, will do to secure the general causal certainty (certitudo ordinis) of divine providence, but it is insufficient in the case of predestination. He ignores the objection that the proximate, created cause in the realization of predestination is not a natural, but a voluntary cause, viz. free will, which, by its nature, is not a contingent cause inclined to one alternative (contingens ut in pluribus) and failing only occasionally, but one that is completely indifferent (ad utrumlibet). Instead, he focuses on the doctrinal point that predestination does not regard humanity in general, but the final end of each individual person.57 Furthermore, in contrast with the discussion in the commentary on the Sentences, Thomas acknowledges that it is theologically unacceptable to state that the certainty of predestination only depends on the certainty of God’s knowledge. 58 His answer to the problem is that at a particular instance human free will may fail and turn away from God and his final destination, but that God offers so many opportunities that, in the end, he is successful (so to say) and the person reaches the predestined goal.59 Thomas’ mature view consists in pointing out the transcendence of God’s causality: God causes being as such, including its modal differences. Thomas restricts the validity of the Aristotelian rule that an effect does not have the modal status of its remote, first cause, but of its proximate, secondary cause, to ordered created causes, and he no longer thinks that it is sufficient to explain the relation between divine and created causation. For the rule suggests that divine causation is a kind of mediated, indirect emanation which happens by blind necessity of nature and not intentionally. But this, Thomas says, is not in accordance with Christian faith. God is not a first, necessary cause in the way the sun is. First, God’s causal activity itself is not necessitated by his nature, but is voluntary. Secondly, the efficacy of his causal activity exceeds that of any created cause: it not only causes the effect to be, but also the mode in which it exists, viz. contingently or necessarily.60 It does not destroy creaturely causality, but sustains it. God causes necessary created causes to bring forth necessary effects and contingent causes to produce contingent effects. Elsewhere, Thomas states that God’s will is beyond the order of contingency and necessity, for it is the very cause of being as such, and, hence, of the modal order.61 We may use different models, derived from our created reality, to illustrate the relation between divine and created causation: the Aristotelian models of first and secondary causes, remote and proximate causes, principal and instrumental causes, or the Platonic idea of participation. 62 But none of them captures adequately or expresses univocally the transcendence of the divine cause.63 What does Thomas mean when he says that God causes being, and how does that relate to created causes?64 Being, Thomas states, is the common effect of every cause.65 Created causes, however, are limited in their essence, and, hence in their power. Their causal agency always presupposes something upon which to act, and is confined to the categorical determinations of being: substantial and accidental being.66 Only God, who is subsistent being itself, causes being 12 as such (ens inquantum ens), and to him belongs exclusively that causal act that does not presuppose anything, viz. creation out of nothing. Within the existent created order, creatures do have their proper causal role: they cause something to be this, or to be such. But this created causation is not independent from God. While creatures cause something to be this or such, they do not cause something to be this or such. This is what Thomas means when he says that every created agent causes or gives being, not by their own power, but “in virtue of the divine power” (in virtute divina).67 One should not understand this as if God gives a kind of generic being, while created causes, subsequently, determine and particularize it.68 For Thomas, “to be” is not a univocal, generic notion, but “the actuality of each thing,” “more intimate and deeper than anything else.”69 As each cause operates at a different level, the divine one on the transcendental level of being itself, and the created one on the level of its categorical determinations, there is no competition between the two, nor mutual exclusion: “by keeping reality in being … God immediately operates within every agent, without excluding the operation of the will and of nature.”70 5. Conclusion At the beginning of this paper, it was asked (1) how Thomas Aquinas argues for the compatibility of God’s foreknowledge, providence and predestination with human freedom, and (2) whether his arguments are cogent. With regard to the first question, I have tried to show that Thomas distinguishes two problems. The first one has to do with the infallibility of God’s foreknowledge of the future. The other one with the irresistible efficacy of God’s will as the First Cause. Providence, when understood not only as God’s eternal plan to bring creatures to their destination, but also as the execution of that plan in time, presupposes conceptually both God’s foreknowledge and divine causation. The same goes for predestination, which Thomas calls “a part of divine providence”, and which concerns the supernatural end of rational creatures. It is important to note that this distinction does not reflect any real distinction in God, but stems from the special semantics of the divine names, which differ only conceptually. The cogency of the arguments with which Thomas solves both problems, must be assessed within the framework of his negative theology. It is because of God’s incomprehensible, eternal mode of being that events which are future and contingent, and hence indeterminate, in themselves and in relation to us, are said to be present and determinate in relation to God. Likewise, God’s incomprehensible act of giving being as such, including its modal qualifications, allows us to say that the Creator sustains the causal action of creatures and gives being to their effects in accordance with the necessity or contingency of the secondary causes. ‘Presence’ and ‘causation’ are said analogously of the Eternal One and of the Creator, and signify modes of presence and of causation that elude our grasp. In his discussions on the compatibility of God’s foreknowledge, providence, and predestination with human freedom, Thomas develops only the general conceptual and linguistic framework within which the theologian has to think and speak. For a more concrete and detailed analysis of human freedom, we have to turn to (1) Thomas’ anthropology, where he deals with the human will as distinguished from natural contingent causes, and to (2) his theology of sin and grace, where he discusses our deliverance from evil and our new life in community with the Triune God. 13 1 Quotations from Aquinas in this paper are taken from the Leonine edition. For the works not yet published in this edition, I follow the Marietti editions. When quoting from (recent) Leonine volumes, I shall give the Marietti numbers for convenience sake, but use the Leonine orthography. 2 A selection of the most important texts is: I Sent. 38, 1, 5; 40, 3; De Ver. 2, 12; 6, 3; 23, 5; ScG I, 67; ScG II, 30; ScG III, 72-75; ST I, 14, 13; 19, 8; 22, 4; Meta. VI, 3; Periherm. I, 13-14. I have discussed these texts in more detail in my Free Creatures of an Eternal God. Thomas Aquinas on God’s Infallible Foreknowledge and Irresistible Will (Louvain: Peeters 1996). 3 Cf. H. Denzinger, Enchiridion symbolorum, definitionum et declarationum de rebus fidei et morum, bilingual edition Latin-German, ed. P. Hünermann (Freiburg i. Br. etc.: Herder, 1991), no. 1997 and 1997a (pp. 1610-11). 4 Interest in Molinism revived in the mid-1970s by the work of Alvin Plantinga. Some recent studies are: Eef Dekker, Middle Knowledge (Louvain: Peeters, 2000) and Thomas P. Flint, Divine Providence: the Molinist Account (Ithaca N.Y: Cornell U.P., 1998). 5 Cf. Bañez, Scholastica Commentaria in Primam Partem Summae Theologiae S. Thomae Aquinatis (Madrid: F.E.D.A., 1934), pp. 350-51: “… si Deus per impossibile cognosceret tantum futura contingentia, prout habent esse in suis causis, et non prout sunt praesentia in aeternitate, ejus cognitio esset certa et infallibilis [p. 350] … Deus cognoscit futura contingentia in suis causis particularibus, quatenus ipsae causae particulares subjiciuntur determinationi et dispositioni divinae scientiae et voluntatis, quae est prima Causa.” 6 Already in 1910, the Dominican Thomist A.-D. Sertillanges objected to Bañezianism as a distortion of Thomas’ own position. See his S. Thomas d’Aquin (Paris: Alcan, 1925), vol. 1, pp. 265-67. But prominent Neo-Thomists like G. Manser and, in particular, R. Garrigou-Lagrange defended vigorously the praemotio physica. More recent criticisms of Bañezianism are: Kathryn Tanner, God and Creation in Christian Theology. Tyranny or Empowerment? (Oxford: Blackwell 1988), pp. 141-152 and Brian J. Shanley, “Divine Causation and Human Freedom in Aquinas,” American Catholic Philosophical Quarterly 72 (1998): 99-122. 7 ST I, 13, 5. 8 I Sent. 22, 1, 2, ad 4: “… sed hoc quod in modo significandi importetur aliqua imperfectio, quae Deo non competit, non facit praedicationem esse falsam vel impropriam, sed imperfectam.” See also the contribution of David Burrell in this volume. 9 I Sent. 37, 2, 2: “Omnis autem relatio quae fundatur super aliquam operationem in creaturas procedentem, non dicitur de Deo nisi ex tempore, sicut dominus et creator et huiusmodi; quia huiusmodi relationes actuales sunt, et exigunt actu esse utrumque extremorum.” Cf. also, e.g., ST I, 13, 7 ad 3 and 14,15 ad 1. 10 ST I, 46, 2, ad 1: “… causa efficiens quae agit per motum, de necessitate praecedit tempore suum effectum … Sed si actio sit instantanea et non successiva, non est necessarium faciens esse prius facto duratione.” See also ST I, 45, 2, ad 3 and 104, 1, ad 4. Aquinas’ usual example of instantaneous causation is illumination. He did not know yet that light travels at a limited speed. 11 Cf. ST I, 45, 3, ad 1: “… creatio active significata significat actionem divinam, quae est eius essentia, cum relatione ad creaturam.” 12 I Sent. 38, 1, 1: “… scientia secundum rationem scientiae non dicit aliquam causalitatem, alias omnis scientia causa esset” and De Ver. 2, 14: “… scientia inquantum scientia, non dicit causam activam … a scientia nunquam procedit effectus nisi mediante voluntate, quae de sui ratione importat influxum quemdam ad volita.” 13 De Ver. 23, 4, ad 13: “Cum vero dicitur Deum velle aliquid, non significatur illud aliquid inesse Deo, sed tantum modo importatur ordo ipsius Dei ad illius constitutionem in propria natura” and ibid. ad 6: “… illa quae important comparationem Dei ad exitum creaturarum in esse, sicut velle, creare, et huiusmodi …” 14 De Ver. 23, 4, ad 15: “… voluntas ad volitum habet duplicem respectum: primum quidem habet ad ipsum in quantum est volitum; secundum vero habet ad idem, in quantum est producendum in actu per voluntatem; et hic quidem respectus praesupponit primum. Primo enim intelligimus voluntatem velle aliquid; deinde ex hoc ipso quod vult illud, intelligimus quod producat ipsum in rerum natura, si voluntas sit efficax.” 15 De Ver. 23, 4, ad 15: “Primus ergo respectus divinae voluntatis ad volitum non est necessarius absolute, propter improportionem voliti ad finem … Sed secundus respectus est necessarius propter efficaciam divinae voluntatis; et exinde est quod de necessitate sequitur, si Deus vult aliquid voluntate beneplaciti, quod illud fiat.” The ‘will of pleasure’ (voluntas beneplaciti) is a technical Scholastic term. It signifies God’s will in the proper sense of the word. It is contrasted with the ‘will of the sign’ (voluntas signi), which is metaphorically called ‘will’ and signifies God’s commandments: cf. De Ver. 23, 3. 16 Cf. especially De Ver. 5, 1 and 6, 1. 17 Cf. I Sent. 40, 1, 1; ST I, 22, 1, ad 2 and 23,2. 18 I Sent. 40, 3, 1, ad 5: “Unde cum dicitur Deus praescit hoc, non intelligitur quod sit futurum respectu divinae scientiae, sed respectu huius temporis in quo profertur.” Cf. also De Ver. 2, 12 (in fine); ScG I, 67, 9. 19 I Sent. 41, 1, 5, ad 1: “[C]um dicimus, Deus praescivit Christum moriturum, designatur respectus ad futurum; unde quando desinit esse futurum, Deus iam non praescit illud …” Cf. also I Sent. 35, exp. txt.; De Ver. 6, 1, ad 6. 20 I Sent. 38, 1, 5: “Contingentia enim videtur duplici ratione effugere divinam cognitionem. Primo propter ordinem causae ad causatum. Quia causae necessariae et immutabilis videtur esse effectus necessarius; unde cum scientia Dei sit causa rerum, et sit immutabilis, non videtur quod possit esse contingentium. Secundo propter ordinem scientiae ad scitum; quia cum scientia sit certa cognitio, ex ipsa ratione certitudinis etiam exclusa causalitate, requirit certitudinem et determinationem in scito, quam contingentia excludit; et quod scientia ex ratione certitudinis suae requirat determinationem in scito, patet in scientia nostra, quae non est causa rerum, et in scientia Dei respectu malorum. ” 21 Cf. I Sent. 40, 3, 1; ST I, 23, 6; De Ver. 6, 3; Periherm. I, 14, 192; De Malo 16, 7, ad 15; ScG III, 95, 9 and 16; Comp. Theol. I, 140; Quodl. 12, 3. 22 De Ver. 6, 3: “… duplex est certitudo: scilicet cognitionis et ordinis. Cognitionis quidem certitudo est, quando cognitio non declinat in aliquo ab eo quod in re invenitur, sed hoc modo extimat de re sicut est; et quia certa extimatio de re praecipue habetur per causam rei, ideo tractum est nomen certitudinis ad ordinem causae ad effectum, ut dicatur ordo causae ad effectum esse certus, quando causa infallibiliter effectum producit.” 23 ScG I, 67, 2: “Contingens enim certitudini cognitionis non repugnat nisi secundum quod futurum est, non autem secundum quod praesens est. Contingens enim, cum futurum est, potest non esse … Ex quo autem praesens est, pro illo tempore non potest non esse.” 24 What I call ‘temporal necessity’ is Thomas’ ‘quaedam necessitas’ in ST II-II, 49, 6: “Praeterita autem in necessitatem quandam transeunt: quia impossibile est non esse quod factum est. Similiter etiam praesentia, inquantum huiusmodi, necessitatem quandam habent: necesse est enim Socratem sedere dum sedet.” Cf. I Sent. 42, 2, 2 and 44, 1, 4; ScG II, 25, 15; ScG III, 86, 6; Eth. VI, 2, 1138; Resp. de art. 108, 82, 1; De Coelo I, 29, 10. The Scholastics also called it ‘necessitas per accidens’ or ‘necessitas ut nunc’. 25 The metaphor of the moving tip of the pen also illustrates the difficulties in imagining and conceptualizing time: the metaphor itself includes what it is supposed to clarify, viz. time. 26 The so-called ‘old’ tenseless theory held that the tensed and the tenseless phrase have exactly the same meaning. According to the ‘new’ tenseless theory, semantic equivalence is not necessary; it suffices that the tenseless phrase expresses effectively the extensional truth conditions or facts. 27 A clear introduction in contemporary philosophy of time is: Quentin Smith and L. Nathan Oaklander, Time, Change and Freedom: An Introduction to Metaphysics (London: Routledge, 1995). 28 A solution to the problem of foreknowledge along these lines, has been labelled the ‘Ockhamist’ solution. It hinges upon the notion of counterfactuality: God’s foreknowledge would have been different, if I would have made a different choice than I actually shall. It considers knowledge to be a ‘future-infected soft fact’, depending semantically on what will actually happen. This line of reasoning is followed, among others, by William L. Craig. See for instance his Divine Foreknowledge and Human Freedom. The Coherence of Theism: Omniscience (Leyden: Brill, 1991). In this study Craig defends the compatibility of the Ockhamist solution with a tensed theory of time. Later he advocated the tenseless view in “The Tensed vs. Tenseless Theory of Time: A Watershed for the Conception of Divine Eternity,” in Questions of Time and Tense, ed. by R. Le Poidevin (Oxford: Clarendon 1998): 221-250, esp. pp. 246-48. Aquinas disagrees with the counterfactuality solution: I Sent. 38, 1, 5, ad 4; De Ver. 2, 12, ad 7; ST I, 14, 13 ad 2. 29 Boethius, De Consolatione Philosophiae V, 6 (London: Loeb edition, 1973), 422-435. Like Boethius, Aquinas also uses the Neoplatonic metaphor of the center of a circle, to which the whole circumference is equally present. 30 A. Kenny, “Divine Foreknowledge and Human Freedom” in Aquinas: a Collection of Critical Essays, ed. A. Kenny (London: McMillan 1969): 255-270. Kenny concludes: “But on St. Thomas’ view, my typing of this paper is simultaneous with the whole of eternity. Again, on his view, the great fire of Rome is simultaneous with the whole of eternity. Therefore, while I type these very words, Nero fiddles heartlessly on” (p. 264). The same objection had already been raised by the late- thirteenth-century Dominican William Peter of Godin: cf. Maarten Hoenen, Marsilius of Inghen: Divine Knowledge in Late Medieval Thought (Leyden: Brill 1993), 169-170. 31 Stump and Kretzmann, “Eternity,” Journal of Philosophy 78 (1981): 429-458. 32 For a survey of the debate and a critical evaluation see Craig, “Tensed vs. Tenseless Theory” (mentioned in note Error: Reference source not found above): 232-240. 33 Craig, “Tensed vs. Tenseless Theory”: 235, 239. 34 The analogy with the present moment of time encompassing all of space is used by Anselm: De Concordia praescientiae et praedestinationis et gratia Dei cum libero arbitrio I, 5. 35 Cf. ScG I, 66, 8; ST II-II, 95, 1. In De Rat. Fid. 10 Thomas uses the metaphor of the circle: “… in circulo, cuius centrum … aequaliter respicit omnes circumferentiae partes, et omnes sibi sunt quodammodo praesentes, quamvis una earum alteri non sit praesens … Deus omnia quae nobis sunt vel praeterita vel praesentia vel futura, quasi praesentia inspiciens, infallibiliter et certitudinaliter cognoscit … Deus autem de excelso suae aeternitatis per certitudinem videt quasi praesentia omnia quae per totum temporis decursum aguntur.” Also Boethius uses ‘quasi’ once: “… omnia quasi iam gerantur in sua simplici cognitione considerat” (De Cons. Phil. V, 6, 65-66). 36 Among others: William L. Craig, “Was Thomas Aquinas a B-Theorist of Time?” New Scholasticism 59 (1985): 475-483; Delmas Lewis, “Eternity, Time and Tenselessness,” Faith and Philosophy 5 (1988): 72-86. John Yates stresses that God’s knowledge is not dependent on observation: The Timelessness of God (Lanham MD: Univ. Press of America, 1990), esp. pp. 221-232. Craig has argued that Scotus criticized Aquinas for endorsing a tenseless view: Craig, The Problem of Divine Foreknowledge and Future Contingents From Aristotle to Suarez (Leyden: Brill, 1988), 129-133. 37 Cf. my Free Creatures, pp.101-136, 220-235. A more detailed survey of thirteenth-century tense logic can be found in my “Tense Logic in 13th-century Theology,” Vivarium (in press). 38 Cf. the detailed study by Richard Gaskin, The Sea Battle and the Master Argument. Aristotle and Diodorus Cronus on the Metaphysics of the Future (Berlin: De Gruyter, 1995). 39 Boethius attributes the epistemic reduction to the Stoics. Cf. his second commentary on De Interpretatione III, 9, pp. 192,194, 245 (New York : Garland, 1987, reprint of the Leipzig edition of C. Meiser 1887-90). 40 For example: ST I, 10 (cf. the prologue in ST I, 3) and ScG I, 30, 4. 41 Boethius, De Consolatione Philosophiae V, 6: “Aeternitas est interminabilis vitae tota simul et perfecta possessio” (p. 422). Aquinas’ apophatic reading of the Boethian definition is pointed out, among others, by Yates, Timelessness, pp. 39-40 and Brian Shanley, “Eternity and Duration in Aquinas,” The Thomist 61 (1997): 525-48. 42 Stump and Kretzmann, “Eternity” (cf. note Error: Reference source not found above). 43 For a survey of the discussion, see William L. Craig, “The Eternal Present and Stump-Kretzmann Eternity,” American Catholic Philosophical Quarterly 73 (1999): 521-536. 44 Cf. I Sent. 8, 2, 1, ad 6, in commenting on Boethius’ definition: “[D]uratio dicit quamdam distensionem ex ratione nominis: et quia in divino esse non debet intelligi aliqua talis distensio, ideo Boetius non posuit durationem, sed possessionem.” Cf. also De Coelo I, 21, 9. On the other hand, in I Sent. 19, 2, 1 (cf. II Sent. 2, 1, 1): “[T]ria praedicta nomina [sc. aeternitas, aevum, tempus, H.G.] significant durationem quamdam. Duratio autem omnis attenditur secundum quod aliquid est in actu.” Eternity is also called duratio Dei in ScG II, 35, 6 and II Sent. 1, 1, 5, ad 7. 45 De Causis 30 (no. 450 in the Marietti edition of 1972): “Et differunt hae perpetuae durationes tripliciter: primo quidem quia perpetuitas aeternalis est fixa, stans immobilis; perpetuitas autem temporalis est fluens et mobilis … Secundo, quia perpetuitas aeternalis est tota simul quasi in uno collecta; perpetuitas autem temporalis habet successivam extensionem secundum prius et posterius quae sunt de ratione temporis. Tertio, quia perpetuitas aeternalis est simplex, tota secundum seipsam existens, sed universalitas sive totalitas perpetuitatis temporalis est secundum diversas partes sibi succedentes.” 46 Stump and Kretzmann, “Eternity, Awareness and Action,” Faith and Philosophy 9 (1992): 463-82, esp. pp. 464-65 and 468-69. 47 ScG I, 66, 8: “Aeterno autem non potest aliquid praesentialiter coexistere nisi toti: quia successionis durationem non habet.” 48 Cf. Albert Michel, “Science de Dieu,” Dictionnaire de théologie catholique 14.2 (Paris: Letouzey, 1941): 1598-1620; and the references in Delmas, “Eternity” p. 85 note 21. In ST I, 14, 13 Thomas says that all of time is present to God “non solum ea ratione qua habet rationes rerum apud se praesentes, ut quidam dicunt …”. The reference is probably to Bonaventure. See also I Sent. 38, 1, 5 (in fine). 49 ST I, 89, 7, 3: “… futura, quae distant secundum tempus, non sunt entia in actu. Unde in seipsis non sunt cognoscibilia, quia sicut deficit aliquid ab entitate, ita deficit a cognoscibilitate.” Cf. also De Malo 16, 7. In other texts, Thomas states that sometimes we cannot know certain things due to a deficiency of our cognitive powers, as in the case of the mystery of the Trinity, while other things cannot be known (by any cognitive faculty) because of a deficiency from their side, like future contingents: ST II-II, 171, 3; De Ver. 12, 2 and Ad Rom. 12, 2, no. 978. 50 Cf. John F. Wippel, “Thomas Aquinas and the Axiom ‘What is received is Received According to the Mode of the Receiver’” in A Straight Path. Studies in Medieval Philosophy and Culture: Essays in honor of Arthur Hyman, ed. by R. Link-Salinger e.a. (Washington D.C.: Cath. Univ. of America Press, 1988), 279-289. 51 Cf. I Sent. 38, 1, 3, ad 3: “ … temporalia intemporaliter cognoscit” and Periherm. I, 14, 195: “… omnino eternaliter sic videt unumquodque eorum que sunt in quocunque tempore.” Nor does God know by forming tenseless propositions. Such propositions imply a tenseless view on time. 52 Cf. De Ver. 2, 12: “Difficultas autem in hoc accidit eo quod divinam cognitionem significare non possumus nisi per modum nostrae cognitionis consignificando temporum differentias.” In the answer to the seventh objection, Thomas states: “… et ideo magis esset dicendum ‘si Deus scit aliquid, hoc est’ quam ‘hoc erit’.” Thomas’ counterfactual phrase already indicates that ‘hoc est’ is also imperfect, for the future contingent does not exist now. Cf. also Ep. ad Bern.: “Eo ergo modo ab eterno prescivit hunc tali tempore moriturum si modo nostro loquimur, cum tamen eius modo dicendum esset, videt eum mori …” 53 ScG II, 30, 1: “Licet autem omnia ex Dei voluntate dependeant sicut ex prima causa, quae in operando necessitatem non habet nisi ex sui propositi suppositione, non tamen propter hoc absoluta necessitas a rebus excluditur, ut sit necessarium nos fateri omnia contingentia esse; quod posset alicui videri, ex hoc quod a causa sua non de necessitate absoluta fluxerunt.” 54 Cf. Meta. VI, 3, 1210. Thomas’ views on coincidence in natural causation receive little attention, partly because they are interwoven with outdated scientific (cosmological) theories, partly because ‘coincidence’ is also a problematic notion in modern science. Still, Thomas’ texts contain valuable discussions on the concept of ‘cause’ and on the compatibility of natural coincidence with divine providence: cf. my Free Creatures, pp. 281-89. 55 The development is sketched by Bernard McGinn in his “The Development of the Thought of Thomas Aquinas on the Reconciliation of Divine Providence and Contingent Action,” The Thomist 39 (1975): 741-52. 56 I Sent. 38, 1, 5: “Sed adhuc manet dubitatio maior … quia causa prima necessaria potest simul esse cum defectu causae secundae, sicut motus solis cum sterilitate arboris; sed scientia Dei non postest simul stare cum defectu causae secundae. Non enim postest esse quod Deus sciat simul hunc cursurum, et iste deficiat a cursu; et hoc est propter certitudinem scientiae et non propter causalitatem eius.” As we say earlier (cf. note Error: Reference source not found above), Thomas discusses divine knowledge both as cause and as knowledge in this text. Cf. also II Sent. 35, 1, 5, exp. txt. where Thomas says that what happens rarely in nature is fortuitous or casual compared to its proximate cause, but not “in ordine ad causam primam, cuius praesentiam nihil praeterfugit.” 57 De Ver. 6, 3: “Et sic nihil potest deficere a generali fine providentiae, quamvis quandoque deficiat ab aliquo particulari fine. Sed ordo praedestinationis est certus non solum respectu universalis finis, sed etiam respectu particularis et determinati … Nec tamen hoc modo est certus ordo praedestinationis respectu particularis finis, sicut erat ordo providentiae: quia in providentia ordo non erat certus respectu particularis finis, nisi quando causa proxima necessario producebat effectum suum; in praedestinatione autem invenitur certitudo respectu singularis finis; et tamen causa proxima, scilicet liberum arbitrium, non producit effectum illum nisi contingenter.” On the expression ‘certitudo ordinis’, see note Error: Reference source not found above. 58 De Ver. 6, 3: “Non enim potest dici quod praedestinatio supra certitudinem providentiae nihil aliud addat nisi certitudinem praescientiae … Sic enim non diceretur praedestinatus differre a non praedestinato ex parte ordinis, sed tantum ex parte praescientiae eventus. Et sic praescientia esset causa praedestinationis, nec praedestinatio esset per electionem praedestinantis; quod est contra auctoritatem Scripturae et dicta sanctorum.” 59 De Ver. 6, 3: “Tamen in eo quem Deus praedestinat, tot alia adminicula praeparat, quod vel non cadat, vel si cadit, quod resurgat, sicut exhortationes, suffragia orationum, gratiae donum, et omnia huiusmodi …” 60 De Ver. 23, 5 is one of the first texts that represents Thomas’ mature view. He begins by quoting the Aristotelian rule, adduced by ‘quidam’ (including the young Thomas himself!). After having given the objections, he concludes: “Et ideo oportet aliam principalem rationem assignare contingentiae in rebus, cui causa praeassignata [Aristotle’s rule, H.G.] subserviat … Voluntas autem divina est agens fortissimum, unde oportet eius effectum ei omnibus modis assimilari, ut non solum fiat id quod Deus vult fieri … sed ut fiat eo modo quo Deus vult illud fieri, ut necessario vel contingenter … Et sic non dicimus quod aliqui divinorum effectuum sint contingentes solummodo propter contingentiam causarum secundarum, sed magis propter dispositionem divinae voluntatis, quae talem ordinem rebus providit.” Cf. also Quodl. 11, 3 written around Easter 1259, in the same period as De Ver. 23. 61 Cf. Periherm. I, 14, 197: “Nam voluntas divina est intelligenda ut extra ordinem entium existens, velut causa quedam profundens totum ens et omnes eius differencias; sunt autem differencie entis possibile et necessarium, et ideo ex ipsa voluntate divina originatur necessitas et contingencia in rebus, et distinctio utriusque secundum rationem proximarum causarum.” All effects of created causes depend on God’s will, the “prima causa que transcendit ordinem necessitatis et contingencie” (ibid.). Discussing the efficacy of divine providence in Meta. VI, 3, 1222, Thomas says: “Sed necessarium est effectus esse contingenter, vel de necessitate. Quod quidem singulare est in hac causa, scilicet in divina providentia. Reliquae enim causae non constituunt legem necessitatis vel contingentiae, sed constituta a superiori causa utuntur.” Cf. also ST I, 19, 8; De Malo 16, 7, ad 15; ScG III, 94, 10; Quodl. 12, 3, ad 1. 62 Cf. for example De Pot. 3, 7 where Thomas uses all of these models. 63 On the apophatic character of Thomas’ account of divine causation, see Shanley, “Divine Causation” (cf. note Error: Reference source not found above). 64 For a more detailed account of this question see: Rudi te Velde, Participation and Substantiality in Thomas Aquinas (Leyden: Brill 1995), 160-83; John F. Wippel, “Thomas Aquinas on Creatures as Causes of Esse,” International Philosophical Quarterly 40 (2000): 197-213. 65 ScG III, 66, 3: “… esse sit communis effectus omnium agentium, nam omme agens facit esse actu”; De Pot. 7, 2: “Omnes autem causae creatae communicant in uno effectu qui est esse … Conveniunt ergo in hoc quod causant esse.” Cf. also Meta. V, 1, 751; ScG III, 67, 1; Phys. II, 10, 15. 66 Cf. ScG II, 21, 4: “Alia vero agentia non sunt causa essendi simpliciter, sed causa essendi hoc, ut hominem vel album”; De Pot. 3, 1: “… nulla earum est activa entis secundum quod est ens, sed eius entis secundum quod est hoc ens, determinatum in hac vel illa specie.” 67 Cf. ScG III, 66, 1: “… omnia inferiora agentia non dant esse nisi inquantum agunt in virtute divina”; De Pot. 3, 1: “… nulla res dat esse, nisi inquantum est in ea participatio divinae virtutis.” 68 Sometimes Thomas’ wordings might suggest this. Cf. De Pot. 3, 1: “Causalitas vero aliorum quae ad esse superadduntur, vel quibus esse specificatur, pertinet ad causas secundas …” and ScG III, 66, 6: “Secunda autem agentia, quae sunt quasi particulantes et determinantes actionem primi agentis, agunt sicut proprios effectus alias perfectiones, quae determinant esse.” 69 Cf. e.g. ST I, 4, 1, ad 3: “… ipsum esse est actualitas omnium rerum” and 8, 1: “Esse autem est illud quod est magis intimum cuilibet, et quod profundius omnibus inest.” 70 De Pot. 3, 7: “… tenens rem in esse … ipse in quolibet operante immediate operetur, non exclusa operatione voluntatis vel natura.”