http://www.romancatholicism.org

 

 

Grace and Free Will in Arnauld

 

 

 

 

Elmar J. Kremer

 

(This essay outlines Jansen’s theory of grace and explains how Arnauld maintained an Augustinian theology while avoiding the third proposition attributed to Jansen. He was a theologian and philosopher of extraordinary authority in the seventeenth century and was known as the Great Arnauld by his contemporaries.)

 

 

Antoine Arnauld (1612-1694)

 

 

It is not surprising that Arnauld should have devoted much time and energy to the problem of how the doctrines of original sin and grace are consistent with belief in free will. The problem had been central to Christian theology at least since Augustine, and had more recently become agitated by the Protestant reformation and the Catholic counter-reformation.1 Arnauld’s dealings with the problem were complicated by the fact that they began as part of his defence of Jansen which he undertook in the early 1640s and which occupied much of his time and attention for the next forty years.2

 

 However, about 1684, under the influence of Aquinas, Arnauld came to reject several key tenets of Jansen’s position, and from that time until his death in 1694, he developed a position of his own, which he said he derived from Aquinas’s Summa Theologiae.3 It is with Arnauld’s position in this late period that I am primarily concerned. I will approach it by setting forth Jansen’s position and then explaining in what ways Arnauld departed from Jansen after 1684.

Notes

 

1 In 1547, the Council of Trent declared it de fide that neither original sin nor grace removes freedom of will. See the ‘Decree on Justification’ of the sixth session of the Council, chapter 1 and 5 and canon 4 and 5.

 

 2 Arnauld undertook to defend Jansen, after the posthumous publication of Augustinus in 1640, at the request of his mother’s confessor and Jansen’s old friend, Jean Duvergier, the Abbé de Saint-Cyran. The beginnings of Arnauld’s work on the problem of sin, grace, and free will are described in N. Larriere, Vie de Messire Antoine Arnauld, vol. 43 of Oeuvres de Messire Antoine Arnauld, p. 36ff. (See n. 4 below.)

 

 3 In a letter written about two years before his death, Arnauld recommended ‘a little Latin text’ (un petit ecrit latin) on freedom that he had written some time before, and asked that it be read ‘independently of what I wrote at another time in the Apologies for Jansen; for I was then obliged to defend him in accordance with his principles.’ He adds that ‘it was only seven or eight years ago that I had the occasion to examine thoroughly the true opinion of St Thomas and I realized that the texts usually cited from his commentary on the Master of the Sentences and from his other works prior to his Summa did not agree with what he teaches on this matter in the latter work, which is his masterpiece and hence should be taken as definitive.’ (Letter to Vuillaret, 21 June 1692). Unless otherwise indicated, references to Arnauld are to Oeuvres de Messire Antoine Arnauld, 43 vols (Lausanne: Sigismond D’Arnay et Cie, 1775-83). References to this collection will be made by giving the volume and page numbers. (The letter just cited is found at 3:498.)

 

 This letter indicates that Arnauld had begun reworking his views on free will, under the influence of Aquinas, by 1684. In a work written in 1690, Arnauld says that he wrote the ‘petit ecrit latin’ on freedom two years earlier, which would put its composition at about 1688. (See n. 27 below.)

1. Jansen on the Compatibility of Original Sin, Efficacious Grace and Free Will

 

 

Using phrases common in the theological tradition, Jansen says that an action is free just in case it is in the agent’s power, and, hence, the agent is master of the action.4 But what does it mean to say that something is in an agent’s power? It seems to be his view that an action is in the agent’s power just in case performing it requires no more on the part of the agent than an act of will.5 He holds that both external voluntary acts and internal acts of will are in our power. But acts of will, unlike any bodily action, are essentially free: ‘For this is the root of all freedom, at least in all actions, whether external or internal, namely that they are in our power [potestate]. Therefore the will is free by its very nature. For it implies a contradiction to say that the will is not free, just as it does to say that in willing we do not will, or that in willing we do not do what we will: that is to say that the will is not the will.’6 He explicitly rejects the views of some ‘more recent’ thinkers who think that an act is free only if, once everything necessary for the performance of the act is given, we can perform it or not perform it.7

 

 It is worth noting that when Jansen says that both acts of will and some external human actions are in our power, he is not suggesting that what these have in common is that they are the results of acts of will, as if behind every act of will there was another act of will. He seems rather to give a component account of action.8 On Jansen’s version, voluntary external actions include acts of will as parts of themselves, so that all such actions, like acts of will themselves, are exercises of the power or faculty of the will.

 

 Now Jansen holds that the will is moved to act by preceding cognitive and appetitive states, and that these include the effects of both original sin and efficacious grace. The appetitive states that precede the act of will proper are not the outcome of deliberation. He calls them ‘delectationes’ (pleasures), and describes them as responses to the person’s cognitions of objects. Delectationes, we might say, are cases of finding this or that object delightful or attractive. These pre-deliberation acts, Jansen says, occur in the will and move the will to deliberate desire: ‘the true cause of free will’s being moved to desire something, is the delectatio that stirs inwardly in the will itself.’9 At any given time, more than one delectatio may stir in a person. In this case the person is determined to will whatever object corresponds to the strongest one.10

 

 Jansen treats both concupiscence, which is the effect of original sin on the will, and efficacious grace as pre-deliberation delectationes.11 Concupiscence is a habitual tendency to love ‘lower things.’ It is a tendency so strong that, unless it is checked by efficacious grace, it makes it impossible to do good or even to avoid sin.

 This corrupt tendency is outweighed by the love of God in those who receive efficacious grace, earned for human beings by Christ.12 Efficacious grace is ‘an inspiration of affection [of love for God] given so that we should do with holy love the things we know [we ought to do].’13 It is not the only sort of grace. Indeed, Jansen sometimes uses the term ‘grace’ in a very general way, to refer to anything God gives a created person that can help the creature to attain its final end, namely, to possess God, the sovereign good, eternally, by knowing Him ‘face to face’ and loving Him. Used in this general way, ‘grace’ refers, among other things, to the creature’s nature as an intelligent being, and all its natural powers.14 But strictly speaking, grace is a divine assistance that goes beyond the natural endowments of any created person, and without which the person cannot attain the above-mentioned end.15

 

 Supernatural grace is subdivided by Jansen into grace of the intellect and grace of the will.16 The latter is in turn divided into habitual and actual grace. Habitual grace of the will is charity, taken as a standing disposition to love God and to choose to perform other acts because of one’s love of God; actual grace of the will, by contrast, is an inner movement of love for God. Jansen applies these distinctions to the grace given to Adam and Eve and the angels (also called ‘the grace of the Creator’), as well as the grace earned by Christ to repair the damage caused by Adam’s sin.17 Finally, efficacious grace, with which our present problem is concerned, is the actual grace of the will earned for human beings by Christ.

 

 There is an important difference, Jansen says, between efficacious grace, on the one hand, and the actual grace given to the angels and the first parents, on the other. The latter only enabled the original parents and the angels to love God and to choose to act for the love of God; whether they did so was decided by their choice, after the grace had been received. Efficacious grace, by contrast, brings it about that its recipient loves God and chooses to act in a meritorious way. Thus, efficacious grace is stronger than the grace given to the first parents and the angels. Efficacious grace is a ‘delectatio victrix,’ which necessarily brings about the good act for which it was given in divine providence. Jansen held that because we post-lapsarian human beings, unlike the first parents before the fall, are affected by concupiscence, we have a greater need for grace, and that the grace of Christ answers to this greater need.18

 

 The distinction between the grace of the Creator and the grace of Christ is closely connected with Jansen’s criticism of Pelagianism and Semi-Pelagianism. Pelagianism is the heretical doctrine that human beings, after the fall of Adam, can perform meritorious actions by their own free will, acting independently of grace. Semi-Pelagianism is a group of heretical doctrines that grew out of Pelagianism and imply that, even it we cannot achieve salvation without supernatural grace, we can contribute to our salvation by the use of our natural powers acting apart from grace.19 To escape completely from these heresies, Jansen maintained, it is not sufficient to say that we, like the angels and our first parents, cannot perform meritorious actions unless we first receive grace. It is necessary to add that our free will can make no contribution to meritorious action except insofar as it is moved by grace, in particular, by a supernatural inspiration of the love of God.

 

 So Jansen thought that when we sin, our will is determined by a pre-deliberate stirring of love for lower things, and that when we perform meritorious acts, our will is determined by a pre-deliberate stirring of the love of God. But he thought that causal determination of this kind did not threaten the freedom of our will because it does not constrain the will. Regarding sin, Arnauld summarizes Jansen’s view as follows: ‘[According to Jansen] freedom is not entirely destroyed in the will of sinners; but only freedom to do good and to abstain from doing evil, which is only one of the two branches of freedom; the other consists in doing evil voluntarily.’20 Regarding meritorious actions, Jansen quotes with approval a text from Aquinas’s Commentary on the Sentences: ‘Even if it [the will] were determined to a single thing, such as loving God, from which it could not refrain, nonetheless it would not lose freedom or praise-worthiness, because it would go toward that thing not under constraint, but spontaneously, and so be master of its act.’21

 

 To this basic compatibilism, Jansen adds one important qualification. It is, he says, taught ‘by Scripture, by St Augustine, by the Fathers, and by the faith of the Church’ that during this life on earth we have a further kind of freedom: ‘Therefore the freedom of man the pilgrim [hominem viatorum] is not only exempt from constraint, but even from immutable voluntary necessity [necessitatis immutabilis voluntariae]; that is, this freedom is indifferent between doing good and doing evil, between acting or not acting, as we freely confess, in accordance with Scripture, St Augustine, the Fathers and Catholic faith.’22 Here he is referring, in part, to the doctrine that the will of a person during the present life is ‘flexible’:23 one who has received grace and hence is leading a meritorious life can always fall out of grace and cease loving God; in that case he would not have had the grace of perseverance. Similarly, a sinner can receive the grace of conversion and begin to love God. But he takes the doctrine to go beyond that and to imply that a person, at any given time during life on earth, has the capacity to choose otherwise than he does choose. So he says that during this life on earth we enjoy freedom of indifference.

 

 However, Jansen rejects the view of unnamed ‘Scholastics,’24 that ‘no matter how free will is attracted by the pleasures of grace or of sin, it can always happen, in the presence of either of those dispositions, that it will the good or the bad.’ This doctrine was rejected by Augustine, he declares, and stubbornly maintained by the Pelagians. According to Jansen, the possibility that someone in the grips of concupiscence should choose what is good, or that someone who receives efficacious grace should choose what is bad, must be understood, ‘not in the sense they [philosophers] call composite, but in the divided sense; that is to say, at the same time that the free choice of the will is filled with the victorious pleasure of grace, which moves it efficaciously, and even when it actually does what is good, the power [potestas] not to do it, and even to sin, is in the will ... And thus, no matter how much the will is pleasantly [suaviter] attracted by grace, it can fail to do what grace makes it do, because it always retains a true power [veram potestatam] not to act, even when it is seized by grace [sub gratia rapiente].’25

 

 Thus, it seems to be Jansen’s view that the will retains the power to do the good even when it is determined by concupiscence to sin, and retains the power to sin even when determined by grace to act meritoriously. It is not obvious at first glance that this is consistent with Jansen’s doctrine that, without grace, a person is unable to do good. There are, however, many cases in which we would say that a person has the power to do something, but is unable to do it.26 Thus, a sighted person who is blindfolded has the power of sight but is unable to see, and a person with functioning legs who is tied to a post has the power of walking but is unable to walk. In a similar way, Jansen holds that, without grace, fallen man retains the power to do good though he is unable to do good, and with efficacious grace retains the power to sin though he is unable to sin. In other words, a fallen man always could act differently than he does if the conditions in which he acts were different – if he had the grace he lacks or if he lacked the grace he has.

4 Cornelii Jansenii, Augustinus, 3 vols (Rothmagi, Sumptibus loannis Berthelin 1643), vol. 3, bk 6, chap. 2, p. 258a (all subsequent references to Jansen are to this work). A book-by-book summary is given in Nigel Abercrombie, The Origins of Jansenism (Oxford: Clarendon Press 1936). A more extensive, and more reliable, summary is contained in the article ‘Jansenism’ by J. Carreyre in the Dictionnaire de théologie catholique, vol. 8, pt 1, pp. 330-448.

 

 5 Jansen cites a formula used repeatedly by Augustine, according to which we have those acts and things in our power ‘quod cum volumus facimus’ (vol. 3, bk 6, chap. 4, 260). It is not easy to translate the formula, but I take it to mean that we have in our power whatever we do (or make act) by willing.

 

 6 Ibid., vol. 3, bk 4, chap. 4, 262.

 

 7 Ibid., vol. 3, bk 6, chap. 4, 260a.

 

 8 This sort of account has recently been developed by Irving Thalberg, Perception, Emotion and Action (New Haven: Yale University Press 1977), and J. Thomson, Acts and Other Events (Ithaca: Cornell University Press 1977).

 

 9 ‘qua intrinsecus in ipsa voluntate mulcetur.’ vol. 3, bk 7, chap. 3, 310a.

 

 10 Ibid., vol. 3, bk 6, chap. 35, 310b.

 

 11 On concupiscence, see Jansen, vol. 2, bk 2, De Statu Naturae Lapsae, chap. 7, 130a: ‘Concupiscence, or libido, or cupiditas, or voluptas, or delectatio, by whatever of these words that evil is called, is nothing other than a certain habitual weight which inclines the soul toward the enjoyment [fruendam] of creatures, or as Augustine says, of inferior things.’

 

12 Concupiscence is not, however, removed. Jansen says that even when one receives efficacious grace, he has in himself ‘that same weight and inclination toward sin which is the perfect power for sinning [peccandi potestas perfectissima]’ (quoted by Arnauld in Seconde apologie, 17:183, from Jansen, vol. 3, bk 8, chap. 20).

 

 13 The phrase, a favourite of both Jansen and Arnauld, is taken from Augustine, Ad Bonif., bk 4, chap. 5: ‘Inspiratio dilectionis ut cognita sancto amore faciamus.’ Jansen and Arnauld both take ‘dilectionis’ to refer to charity or the love of God. For Arnauld’s explication of the phrase, see especially the Seconde apologie, 17:604-5.

 

 14 Jansen, vol. 2, Liber Singularis de Statu Naturae Innocentis, chap. 19, p. 71a.

 

 15 In Arnauld’s words, ‘M. Ypres holds above all that in whatever state a human being is created, he cannot love God, and even less enjoy God forever in glory, except by a true grace, and by a principle infinitely above all the principles of his nature.’ (Seconde apologie pour Jansenius, [hereafter cited as Seconde apologie], 17:146-7). Jansen is called ‘M. Ypres’ after the diocese of which he was bishop. Arnauld is here distinguishing Jansen’s views from those of Michel du Bay, who was condemned by Pius V and Gregory XIII for having held that the state of original innocence of the first parents was a natural state, not a supernatural grace.

 

 16 Jansen, vol. 3, bk 1, chap. 4. Grace of the intellect consists in the external grace of Sacred Scripture and church teaching and the internal grace of understanding and accepting Scripture and church teaching.

 

 17 Jansen, vol. 2, Liber Singularis de Statu Naturae Innocentis, chap. 19, 7 Ib.

 

 18 This distinction is fundamental to Jansen’s interpretation of Augustine, and he buttresses it by frequent quotation from Augustine’s anti-Pelagian writings. Arnauld summarizes the main texts, including the following: The grace of Adam and the angels is a ‘help without which’ meritorious actions are not done (Augustine’s Latin: adjutorium sine quo aliquid non fit), whereas the grace of Christ is a ‘help by which’ meritorious actions are done (Augustine’s Latin: adjutorium quo aliquid fit). The grace of the first parents and the angels was ‘a help such that they could not make use of it if they willed not to, or could make use of it if they willed to, but which did not itself make them will.’ The grace of Christ, by contrast, is, ‘not only a help without which we would be unable to do the good, or persevere in the good, even if we wanted to; but is of such a kind, and so strong, that it makes us want to; it is a grace which gives us not only the power to will what we will, but the willing of what we have the power to will’ (Seconde apologie, 17:169-70). The passages are quoted by Arnauld from Augustine’s De Correptione & Gratia, chaps. 11 and 12. Arnauld’s French translations, which I render into English, are not literal. He gives Augustine’s Latin in footnotes.

 

 19 Jansen distinguishes between Pelagianism and Semi-Pelagianism in vol. 1, bk 1. In an appendix to vol. 3, he tries to show that the views of certain modem thinkers, Molina, Suarez, Vasquez, and Lessius, are parallel to those of the Semi-Pelagians.

 

 20 Seconde apologie, 17:264. The main texts in Jansen are vol. 2, bk 2, chap. 8 and 9. Near the end of chap. 8, Jansen says, ‘without that help [of supernatural grace] the will [arbitrium] cannot be free from servitude to cupidity so as to will the good [liberum a servitute cupiditatis ad volendum bonum]’ (192a).

 

 21 In 3 Sent., disp. 18, a. 2, ad 5. The text is quoted by Jansen in vol. 3, bk 8, chap. 15,361a.

 

 22 Quoted by Arnauld in Seconde apologie, 17:242. He hastens to add that ‘the same Fathers would deny that, speaking generally, human freedom of will consists in this indifference.’

 

 23 Vol. 3, bk 6, chap. 34, 300b.

 

 24 Arnauld takes this to be a reference to ‘the Jesuits, adherents [Sectateurs] to Molina’s opinions’ (17:182).

 

 25 Quoted by Arnauld in the Seconde apologie (17:182) from Jansen, vol. 3 bk 8, chap. 20. I translate Jansen’s Latin, which Arnauld correctly transcribes in a footnote, although it is mistranslated in Arnauld’s French text. On the distinction between the composite and divided senses of modalities (sometimes called the distinction between the de dicto and de re senses), see William Kneale, ‘Modality De Dicto and De Re,’ in Ernest Nagel, Patrick Suppes and Alfred Tarski, eds, Logic, Methodology and Philosophy of Science (Stanford, Calif: Stanford University Press 1962), 626.

 

26 Here there is a problem in translating both Jansen and Arnauld. They sometimes use the Latin ‘potestas’ and the French ‘puissance’ as synonyms for the scholastic term ‘facultas.’ In such contexts, I have translated them by ‘power.’ But in other cases I have translated them by such expressions as ‘ability.’

 

 

2. Arnauld’s Departures from Jansenism

 

 

 

Arnauld never wavered in his acceptance of what he considered the basic doctrines of the Jansenist position: that no creature can perform actions that merit the reward of heaven without supernatural grace; that fallen man cannot avoid sin without the grace of Christ; that the latter grace, in contrast to the grace of the Creator, is efficacious; that efficacious grace consists essentially in an inspiration of the love of God; that neither original sin and concupiscence nor efficacious grace removes free will; that free will makes an essential contribution to meritorious action, but can do so only insofar as it is moved by grace.

 

 However, he departed from Jansen’s account of free will, concupiscence, and efficacious grace in several ways:27

 

1 He rejected Jansen’s doctrine that the only sort of necessity incompatible with free will is necessity of constraint and came to hold that ‘necessity of nature’ is also incompatible with free will.

 

2 He came to hold that freedom of will requires that one be able to will otherwise than one does, and that one determine one’s own act of willing.

 

3 He came to hold that concupiscence renders the will unable to do good ‘only in an improper sense [qu’improprement].’28

 

4 He rejected the doctrine that efficacious grace is a pre-deliberation delectatio, and indeed denied that it is any sort of created entity distinct from the free and meritorious act of will.

 

Let me develop these points one at a time.

 

 1. A close reading of the Summa Theologiae of Aquinas convinced Arnauld that freedom requires not only the absence of constraint, but also the absence of ‘necessity of nature.’ He borrowed the phrase from Aquinas and gives the

following account of its meaning: ‘St Thomas calls it natural necessity when some potency is determined by nature to one act, as when the intellect accepts principles that are self-evident [per se nota] because it is determined by nature to one act, namely the act of accepting them [determinates est per naturam ad illud unum, ut illis adhaereat].29 To say that a potency is determined to one thing is to assume that the potency, to begin with, can be actualized in more than one way, and then the possible actualizations are somehow narrowed down to one.30 When the possible actualizations are narrowed down to one by the nature of the thing whose potency is under consideration, together with the natures of whatever things are acting upon it, then the potency is said to be determined by nature to one act. Strictly speaking, necessity of constraint is a specific form of natural necessity, in which a thing (or, more precisely, some potency in a thing) is determined by nature to an act it would not perform spontaneously, as when a person is determined by an outside force to go where he does not want to go.31 Arnauld, however, uses the phrase ‘necessity of nature’ to refer to necessity of nature that is not necessity of constraint.

 

 Arnauld discusses three cases in which he thinks that human mental powers are determined by nature to a single act, one regarding the intellect and two regarding the will. In general, the intellect can act in two possible ways with regard to a proposition: it can assent or not assent. But if a proposition is present to the intellect as self-evident, the possibilities are narrowed down by nature to one: the intellect cannot but assent.32 So, in this case, the intellect is determined by nature to one act. Similarly, the will can act in either of two ways with regard to an object presented to it as a good: it can ‘love’ or ‘desire’ the object or it can not love or desire it.33 But if the object appears to be only good and to contain no imperfection, the possibilities are narrowed down by nature to one: the will cannot but love or desire the object. Only one object appears to human beings in this way during life on earth, says Arnauld, and that is ‘happiness in general.’ In addition, if one is fortunate enough to see God face to face in heaven, then God will also appear to one in this way. These two circumstances, thinking of happiness in general and having God present face to face, are the only circumstances in which the human will is determined by nature to one act. In every other case, Arnauld claims, it retains the power to desire or not to desire an apparently good object, because the object does not fully satisfy the desire for happiness. This is true even of God, when He is present to someone in this life, not face to face, but by means of an abstract concept.34

 

 Arnauld holds that a potency is determined by nature to one act whenever its act follows upon a preceding event in accordance with a law. Thus, he responds to Malebranche’s statement that the destruction of Jerusalem by the Romans some seventy years after the death of Jesus was ‘a necessary outcome of the laws of nature,’ by saying, ‘It is clear ... that a long series of diverse events, that depended on an infinity of free movements of the wills of men, could have been, and indeed were, regulated by the providence of him who has an almighty power to move the hearts of men, and to direct them as he pleases ... But it cannot be said that it was a necessary consequence of the order of nature, without degrading human beings, making them act like lower animals, and depriving them of their liberty; a position that has been declared anathema by the Council of Trent.’35 He repeats the criticism a few pages later: ‘As I have shown above, to hold that events that depended on an infinity of free volitions of human beings were a necessary consequence of laws of nature is to deprive human beings of their freedom, after the example of Wycliffe.’36 Arnauld assumed that the laws of nature express the natures of the things they govern.37 For this reason he takes Malebranche’s account of these events to imply that the wills of the human agents involved were in every case determined by nature to a single act, and hence were not free.

 

 Arnauld gives two reasons for the claim that when the will is determined by nature to a single act it is not free, in the sense ‘necessary and sufficient for merit and demerit.’38 One reason is that in such a case the person is not ‘master of his act.’39 We have already encountered, in Jansen, the notion that we act freely only when we are masters of our act. As we have seen, Jansen holds that a person is master of his act just in case he performs the act voluntarily, and he explicitly rejects the claim that a person is master of his act only if he could refrain from performing it. Here Arnauld disagrees: he holds that an act of will is free only if the agent can refrain from willing as he does.

 

 Arnauld’s second reason for saying that the will does not act freely when it is determined by nature to one act is that in this case the person does not act ‘by his own initiative and choice.’40 Similarly, he says that the will acts freely only if it is a power ‘which can turn itself to one side or the other.’41 These are causal expressions, implying that the person causes his own act of will, and, indeed, Arnauld quotes with approval Aquinas’s statement, ‘Free will is cause of its own movement.’42 In addition, Arnauld says that when we act freely, we act by ‘determining ourselves [determinando nos ipsos] to what we are not determined by nature,’43 and this seems to imply that when we will freely, we are not causally determined to will as we do.

 

 2. With this second reason Arnauld takes a further step away from Jansen. Jansen says that in this life on earth we can will otherwise than as we do, and hence we enjoy freedom of indifference. Arnauld rejects this terminology. He does not say that the will is indifferent when we can will otherwise that we do. He prefers the phrase, borrowed from Aquinas, that the will is ‘a faculty or power ofopposites.’44 There is a substantive disagreement behind this difference of terminology. Arnauld does not think that Jansen’s freedom of indifference guarantees freedom of will in the sense necessary and sufficient for merit and demerit, because it requires only that the will can be caused to act in more than one way. True freedom of will, according to Arnauld, requires that the will determine itself, by its own initiative, to act as it does.

 

 3. That Arnauld came to differ from Jansen in the way just indicated, is confirmed by differences in their accounts of concupiscence. While Jansen speaks of concupiscence as determining the will, without qualification, to act sinfully, and says that it deprives the will of the freedom to desire the good and even of the freedom to avoid desiring evil,45 Arnauld develops a more complicated position. According to him, concupiscence produces only a ‘voluntary inability’ [impuissance volontaire] to refrain from evil and to do good, not a ‘physical inability’ [impuissance physique]. A voluntary inability is an obstacle to action that one could overcome simply by willing. For example, what renders a miser unable to love God is the miser’s love of money. Furthermore, this love of money is an obstacle that the miser could overcome simply by performing an opposite act of will. So his inability to love God is voluntary. A physical inability, by contrast, is an obstacle that one cannot overcome simply by willing. For example, if a man is blindfolded, his inability to see is physical, and if a man is tied to a post, his inability to walk is physical. Now Arnauld says that concupiscence, in those who sin, is a ‘voluntary inability [to desire what is good], which is an inability only in an improper sense [qui n’est qu’improprement impuissance], because it is only a weakness, a languor, or a sickness of the will, which is strongly attached to what it ought not love, and is thus rendered incapable of loving what it ought to love; although it is in its power [puissance] to love it if it wills [si elle voulait].’46 Here Arnauld preserves the claim that the sinner determines his own will to desire what is evil, by saying that the sinner’s will is not strictly determined by concupiscence. But this seems to imply that we do not need grace to avoid sin, and hence that Arnauld, in order to defend freedom, has fallen into Pelagianism. Arnauld himself raises this objection, but his answer depends on his account of efficacious grace, and so I will not take it up until near the end of the paper.

 

 4. Arnauld not only rejected Jansen’s account of free will and of concupiscence, but also his doctrine that efficacious grace is a ‘delectatio victrix.’ Indeed, he denied that efficacious grace involves any state in the recipient other than a free and meritorious act of will:

 

 the true opinion of St Augustine, St Bernard and St Thomas concerning actual grace, like that of Estius, does not place anything created in the will, between the will of God that he calls uncreated grace and the free movement of the human will that the uncreated grace produces in the human will; which is not to deny that this first free movement of the will produced by the uncreated grace, often serves to produce others, with the help of the same uncreated grace, as St Augustine indicates by this fine definition of grace: For it means that God inspires his love in us in order that we can do good works by means of that love ... But in all of this we see nothing about a flowing quality [qualitas fluens] or an indeliberate act [actus indeliberatus] which M. d’Ypres [i.e., Jansen] said his victorious pleasure consisted in, a point on which he was surely mistaken.47

 

 Arnauld defends his new position in two ways. In Humanae Libertatis Notio, he argues dialectically against the Jansenist view that ‘grace consists in a victorious delectatio, which would be an indeliberate volition, and hence not free.’48 He first says that such a victorious pleasure would have to come after, not before or during, deliberation. He gives the following example: suppose a person who is not very generous meets a beggar who asks for alms. ‘He deliberates, thinks of the commandment of Jesus Christ and gives alms to the beggar for the love of God.’ Now, says Arnauld, those who hold that grace is a victorious delectatio must conclude that three things occurred in the mind of the almsgiver: a deliberation, which occurred in his intellect, followed by a victorious delectatio in his will, followed by ‘a pious volition to give alms which would be an effect of the delectatio.’ Arnauld’s thought seems to be that the supposed victorious delectatio would have to occur after the deliberation is completed, because the delectatio would be overpoweringly strong and hence incompatible with further deliberation. But acts of will are said to be deliberate only because of the deliberation that precedes them in the intellect. Hence, the supposed delectatio victrix would have as good a claim to the title ‘deliberate’ as any other act of will. Jansen, however, is prepared to agree that all deliberate acts of will are free. Hence, the Jansenist doctrine of an indeliberate, and unfree delectatio is incoherent.

 

 In a later text, he gives a different argument against, the doctrine of the delectatio victrix.49 He begins by citing a distinction drawn by Aquinas between the natural love of God and the love of God called ‘charity’ by theologians. The former is nothing other than the desire for happiness, which can be called ‘the love of God’ in that it is a desire that can be filled only by possessing God in the beatific vision. ‘Charity,’ by contrast, is the love of an object under the description of being divine, just as the love, for example, of riches or of honour, is the love of an object under the description of riches or honour, with this difference, that charity is a supernatural grace that God works in us,50 whereas such loves as the love of riches or of honour can be attributed to ‘free will abandoned to itself, in its corrupt state since the sin [of Adam].’ Now although the natural love of God is naturally necessary and hence not free, charity (in this life) is as much a free act of will as the love of any other specific object, like money or honour. But the love of any such specific object is a free act of will. Hence, the love of God in which the created grace in the will consists is a free act of the will. ‘This shows,’ Arnauld concludes, ‘that we ought not posit anything created, non-free and non-deliberate between the mercy of God, which is called uncreated grace, and this free movement of love which He works in our heart.’

 

 Because of these departures from Jansenism, the problem of reconciling efficacious grace with freedom of will was different for Arnauld than it was for Jansen. For Jansen the problem was how to reconcile the claim that efficacious grace is a delectatio victrix that causally determines the will with the claim that the will is free. For Arnauld, by contrast, the problem is how to reconcile the claim that, for any meritorious act of will, God works the act in us with the claim that we determine our will by ourselves, and cause our own act of will.

27 His criticisms of the Jansenist position are stated concisely in ‘Humanae Libertatis Notio,’ the small Latin work referred to in n. 3 above. This work is contained in Causa Arnaldina, ed. by Pasquier Quesnel (1699, apud Hoyoux: Leodici Eburonium), 99-111. A French translation by Quesnel is given in the Oeuvres, 10:614-24, under the title ‘De la liberte de l’homme.’ I will quote the work, under the title ‘Humanae Libertatis Notio,’ giving my translation of the Latin, and page references first to the original Latin and then to the French translation. Arnauld says that this work was based on an analysis of a collection of texts from Aquinas’s Summa Theologiae, presented in Disquisitio utrum, juxta Sanctam Thomam in sua Summa, amor beatificus sit liber ea libertate quam theologi vacant a necessitate, 10:625-40 (hereafter cited as Disquisitio). His later views are also presented in the Instruction sur la grace and Instruction sur l ‘accord de la grace avec la liberté, 10:401-41, and in Ecrit du pouvoir physique, 10: 481-530. Arnauld discusses the matter in several late letters, including: To M. Du Vaucel, 3 Aug. 1691 (3:364-6); To R.P. Macaire, 26 Dec. 1691 (3:417-20); To M. Vuillaret, 21 June 1692 (3:497-9); To M. Du Vaucel, 9 Jan. 1693; To M. Bossuet, July 1693 (3:661-5).

 

 28 Ecrit du pouvoir physique, 10:516.

 

 29 Disquisitio, 10:635.

 

 30 A good account of this Thomistic notion of determinatio ad unum, which was adopted by Arnauld, is given by Elizabeth Anscombe: ‘When we call a result determined we are implicitly relating it to an antecedent range of possibilities and saying that all but one of these is disallowed. What disallows them is not the result itself but something antecedent to the result. The antecedents may be logical or temporal or in the order of knowledge. Of the many – antecedent – possibilities, now only one is – antecedently – possible’ (G.E.M. Anscombe, ‘Causality and Determination,’ Inaugural Lecture, Cambridge University, 1971; reprinted in The Collected Philosophical Pages of G.E.M. Anscombe [Oxford: Basil Blackwell 1981] 3:141). But Arnauld, unlike Anscombe, does not require that determination be antecedent to that which is determined. As we will see, he is prepared to say that the will is self-determining if nothing prior to the person’s willing determines what it is that the person wills.

 

 31 See Disquisitio, 10:628, and compare it with Humanae Libertatis Notio, 104 (10:618).

 

 32 Humanae Libertatis Notio, 110 (10:622). Cf. Instruction sur l’accord de la grace avec la liberté, 10:438.

 

 33 Humanae Libertatis Notio, 100 (10:615). I put ‘love’ and ‘desire’ in raised-eyebrow quotes because they are used in an unusual, or at least now outdated, way by Arnauld, to refer to any act of will. The acts to which he applies the terms include desiring an object in a sense that implies wanting to possess (or stand in some relation to) an object one does not possess; enjoying or liking an object; and wanting to perform an action (desiring to perform an action). Not loving or desiring an object can take the form of loving something incompatible with the object. Thus, Arnauld puts the point that one cannot not desire happiness by saying, ‘the mind is naturally determined to love happiness in general, that is, to will to be happy, in such a way that it cannot not will to be happy, or will not to be happy, or will to be miserable’ (103; 10:617).

 

 34 ‘Since in this life the mind knows God only abstractly ... it is not naturally determined to love Him ... During this life the mind can attend to some of His perfections without thinking of others, and so misers, who think of God only as someone who could give them whatever He wishes, can apprehend God as their enemy when some grave misfortune befalls them, and for that reason hate Him or at least not love Him’ (Humanae Libertatis Notio, 104-5; 10:618-19).

 

 35 Nor, says Arnauld, was Malebranche’s remark a careless slip, for Malebranche had argued two pages earlier that God chose the Jews to be the main prefigurers of Jesus Christ because He foresaw that ‘what would happen to the Jewish people by a necessary consequence of laws of nature had more relation to His plan to foreshadow Christ and his church than what would happen to other nations.’ Malebranche’s remarks are in the Traits de la Nature et de la Grace, Deuxieme Discours, nos. LXI and LXIII. Arnauld’s criticism is given in Reflexions philosophiques et theologiques sur Ie nouveau systems de la nature et de la grace, vol. 1, first published in 1685; 39:301.

 

 36 Ibid., 39:306.

 

 37 Arnauld makes this assumption when he says, regarding Malebranche’s view that volitions are the ‘occasional cause’ of perceptions: ‘That [i.e., the occurrence of our perceptions as a result of our volitions] would be a consequence of our nature ... Therefore, in that case ... God would do no more than execute the laws that he laid down when he instituted our nature’ (Des Vraies et des fausses idees, 38:285; cf. my translation of this text. On True and False Ideas [Lewiston, NY: The Edwin Mellen Press 1990], 107).

 

 38 Humanae Libertatis Notio, 106; 10:620.

 

 39 ‘For our soul would not be free if it were naturally determined to accept everything which is presented to the intellect, and to love everything which is proposed to the will. In that case it could not be said that it moves itself to act and that it is master of its act: Actus sui domina. It could be said that it was drawn along [entrainé] by its object mechanically, so to speak. Therefore reason and will are not free faculties unless they are potentiae ad opposita, powers which can turn themselves to one side or the other. In that alone are they free. St Thomas was right to think that they are not free in those cases where they are determined to a single action by natural necessity’ (Instruction sur l’accord de la grace avec la liberte, 10:438).

 

 40 ‘The common understanding is that only those merit or demerit who do what they do by their own initiative and choice [suopté nutu agere, & ex sua electione]. But who would ever believe that they act by their own initiative and choice when they will what nature determines them to will, and what they are moved [moventur] to will by natural necessity, such as to will to be happy?’ (Humanae Libertatis Notio,

 106-7; 10:620). The root meaning of the Latin word ‘nutus,’ which I have translated ‘initiative’ is ‘nod.’ Quesnel’s French translation has ‘par son propre movement & par son choix.’

 

 41 See n. 39 above.

 

 42 ‘Liberum arbitrium est causa sui motus.’ Quoted in Disquisitio, 10:629, from Summa Theologiae, 1.82.2.

 

 43 Humanae Libertatis Notio, 102; 10:616.

 

 44 ‘The best and most succinct notion one can have of free will, is to say with St Thomas that it is potestas, or facultas ad opposita ... Although that expression seems to mean the same as indifference, it is nevertheless better to use the former than the latter. For the word indifference seems to indicate an equilibrium which is in no way necessary for free will’ (Letter to Bossuet, July 1693; 3:662). The same point is made in letters to R.P. Macaire, 26 Dec. 1691 (3:419); and to Du Vaucel, 3 Aug. 1691, (3:364), 9 Jan. 1693 (3:582), and 9 Mar. 1693 (3:610).

 

 45 See above, n.20.

 

 46 Ecrit du pouvoir physique, 10:516. The phrase ‘it is in [the will’s] power to love [what it ought] if it wills’ does not involve the idea of a second-order volition. Rather it means, ‘it is in [the will’s] power to love [what it ought] simply by so willing.’

 

 47 Letter to Du Vaucel, 8 May 1693; 3:635. Arnauld points out that he had anticipated this position in his Dissertation theologique (published in 1656), pt 3, art. 2 and 4; 20:233-5, 237-9. In article 2 of that early work he divides theologians into those who hold that efficacious grace, considered as a source of good will, consists in the mercy of God together with an ‘inherent form,’ and those who hold that it consists in the mercy of God alone. In article 4 he argues that the latter position implies that ‘efficacious grace pertains not to first act, but to second act.’ He mentions Estius as a theologian of the second sort who recognizes this implication. He does not, however, explicitly declare in favour of Estius’s position or say that Jansen was mistaken on this point. The claim that Jansen’s doctrine is in error is repeated, though without explicit reference to Jansen, in both Humanae Libertatis Notio and Instruction sur l‘accord de la grace avec la liberte.

 

 This fourth departure from Jansen is overlooked by Sleigh when he attributes to Arnauld the view that ‘whenever a human agent S performs a meritorious action A at time t (provided t is after the Fall)... there is a grace g of the efficacious variety such that g occurred in S at t, and the occurrence of g in S brought it about that S chose to do A’ (R.C. Sleigh, Jr, Leibniz & Arnauld: A Commentary on Their Correspondence [New Haven: Yale University Press 1990], 28). This account correctly describes Jansen’s view, but according to Arnauld, efficacious grace, as something received in S, would not be distinct from S’s choice to do A.

 

 48 P. 107; 10:621.

 

 49 Second ecrit sur ce qu ‘enseigne S. Thomas ... de I ‘amour de Dieu ... qu’il appelle naturel, & qu ‘il oppose a I’amour de charite, 10:689.

 

 50 ‘C’est Dieu qui opere ... ce movement d’amour, comme dit Saint Thomas lorsqu’il definit la grace actuelle.’

 

 

 

3. Arnauld’s Reconciliation of Free Will with Efficacious Grace and Concupiscence

 

 

 

To say that God works an act of will in us and that we cause our own act of will did not seem inconsistent to Arnauld, because he accepted the Thomistic theory of primary and secondary causes. According to this theory, everything that has a finite, created cause, also has God as its primary cause. Arnauld took this to mean not only that God causes the created cause to exist, but also that He causes its causing. In this respect secondary causes are like instruments, and hence are often called ‘instrumental causes.’ Suppose a person writes on a blackboard with a piece of red chalk. The person is the principal (though not primary) cause, and the chalk is an instrumental cause, of the writing that appears on the board. Although the chalk’s causal activity is caused by the person who writes with it, none the less the chalk makes a contribution to the outcome: to use Arnauld’s words, the chalk exercises real causal power or efficacy.51

 

 The problem Arnauld faces is whether the proposition that a human agent’s causing of his own act of will is caused by God is consistent with the proposition that it was the human agent who determined which act of will occurred. Suppose it occurs to a person that it would be a good thing to give alms to a beggar. Arnauld’s view is that whether he decides to give the alms or not is up to the person, that the person determines his own will in this regard. The question is whether this is consistent with the proposition that God causes the person’s causing of his act of will. Arnauld asserts that it is:

 

Determination by God by means of grace no matter how efficacious, by which God works in us our volitions and actions [qua Deus operatur in nobis velle & operari], does not hinder our freedom.

 

 For when God works our volition in us [operando in nobis velle]. He brings it about [fecit] that we will in conformity with the kind of thing we are [conformiter nostrae naturae], that is, by determining ourselves [determinando nos ipsos] to what we are not determined by nature. Thus, no matter with what efficacy the mind is moved [moveatur] by God, it acts as master of its very own action; and it wills because it wills, determining itself to all other things by its volition to be happy; and hence it acts freely.52

 

If we think of secondary causes as instruments of God, then Arnauld’s view is that God brings about our meritorious acts of will by using us not as naturally determined instruments, but as free and responsible instruments. An analogy may help make his position clear. Compare God to someone writing with red chalk; a free human agent to the chalk; and the agent’s meritorious act of will to the writing that appears on the blackboard. Then God’s causing of the meritorious act of will by means of the human agent is like the writer’s causing the writing that appears on the blackboard by means of the chalk. But there is this difference: the chalk’s contribution is fixed by its nature; but the human agent’s contribution is not fixed in that way, and is a matter of its own self-determination.

 

 When dealing with this position, it is important to bear in mind that, for Arnauld, causes are primarily things, not events. In particular, a volition is caused, in the basic meaning of ‘caused,’ by God and by the person whose volition it is, not by an event prior to the volition. The same is true of any other act. Thus, the causes of any change in a material thing are, in one way, God, and in another, created things, including persons acting freely and material things acting in a mechanical fashion. Arnauld’s root notion of the cause of an act is that of the origin of the act; the cause of an act is the thing from which the act comes. Thus, when he pursues the question of the ‘efficient cause of our ... perceptions,’53 it turns out to be the question of the origin of our perceptions. The two possible answers he considers are ‘that all our ... perceptions come to us from God, or [that] some of them ... come from ourselves.’54 This notion of a cause allows for a distinction between causation and causal determination: the fact that an act of will comes from God does not in any obvious way imply that it is causally determined by God.55

 

 We are now in a position to deal with the question we earlier postponed: whether Arnauld falls into Pelagianism in his treatment of free will and concupiscence.56 Arnauld raises and answers the question as follows:

 

 But, it will be said, is it not Pelagian to hold that a person has in himself the natural power to love God? It would be if we understood it as the Pelagians do and held that it is in such a way within the force [dans les forces] of free will to love God that we have no need of the medicinal grace of the Saviour in order to love Him in fact [effectivement]. But the natural power to love God can be in our soul, as in fact it is, and be so wounded, weakened and burdened [blessé, affoiblie & appesantie] by a contrary love bending it down towards creatures, like the woman in the Gospel who was bent toward the earth without being able to straighten up, that the soul will never deliver itself from this weight, which is not natural but voluntary, ... if God does not, by His gracious mercy, inspire His love in the soul so as to deliver it from that other love.57

 

As I reported above, Arnauld said that the inability of the sinner to refrain from the love of creatures that is contrary to the love of God, is only improperly called ‘an inability.’ But here he says that it is an inability that the sinner is unable to overcome without the aid of efficacious grace. Are these propositions consistent?

 

 Arnauld’s statement that the sinner has the power to love God does not solve the problem, for when he said that the sinner’s inability to love God is an inability only improperly speaking, he was referring to the sinner’s ability to exercise that power. But the distinction between primary and secondary causality does provide a solution: Arnauld’s view is that the sinner is unable to love God only in an improper sense because there is no secondary cause that strictly speaking prevents him from doing so; concupiscence, in particular, only makes it very difficult to do so. None the less, Arnauld held that a creature is unable to perform any act at all unless it is caused to perform it by God, acting as the primary cause. Now consider the act of turning away from the love of creatures to which we are drawn by concupiscence, and which prevents us from loving God. On the face of it, God, as primary cause, could bring about this act in two ways: by causing in a person an act of will that is within the person’s natural powers as a human being, or by causing in the person a love of God of a kind that is beyond the natural powers of any created being and hence is a supernatural grace. Now Arnauld thought that it was revealed in Scripture and Tradition that God has freely decided to cause post-lapsarian human beings to refrain from the contrary love of creatures only by a supernatural inspiration of the love of God — that is to say, only by efficacious grace. This preserves his anti-Pelagianism, for he can now say that no one can overcome the obstacle of concupiscence and begin to act out of the love of God without efficacious grace. The crucial claim for Arnauld’s reconciliation of grace and free will is that if God does in this way cause the sinner to refrain from that love of creatures, He does so by causing in the sinner an act of will that the sinner also causes, and causes freely.

51 According to Malebranche’s occasionalism, the only true cause is God, the primary cause. On this view, says Arnauld, secondary or created causes ‘have neither force nor power nor efficacy to produce anything at all’ (Reflexions, vol. 1, 39:255). For recent discussion of the doctrine of primary and secondary causes see Philip L. Quinn, ‘Divine Conservation, Secondary Causes, and Occasionalism,’ and Alfred J. Freddoso, ‘Medieval Aristotelianism and the Case against Secondary Causation in Nature,’ in Thomas V. Morris, ed., Divine & Human Action (Ithaca: Cornell University Press 1988).

 

 52 Humanae Libertatis Notio, 102; 10:616. The passage is repeated almost verbatim in Instruction sur I’accord de la grace avec la liberte, 10:439. It echoes Aquinas, Summa Theologiae, I, 83, 1, the third objection and reply:

 

Third Objection: Furthermore, that which is free is cause of itself, as is said in the Metaphysics. Therefore that which is free is not moved by another. But God moves the will. For it is said, ‘The heart of the king is in the hand of God, and he turns it wherever he wills’; and again, ‘It is God who works in us our volition and its accomplishment [qui operatur in nobis velle et perficere]. Therefore man does not have free will.

 

To the third objection it should be said that free will is cause of its own motion: because man through free will moves himself to act. However, it is not necessary for freedom that the being which is free be its own first cause, just as the fact that anything is the cause of something else does not require that it be its own first cause. Thus God is the first moving cause of voluntary causes and natural causes alike. And just as His moving natural causes does not prevent their act from being natural, so his moving voluntary causes does not prevent their actions from being voluntary, but rather brings about such actions in them. For He works in each thing according to what is distinctive of it [secundum ejus proprietatem].

 

 The reply is quoted by Arnauld in Disquisitio, 10:629, but mislabelled I, 82, 2, ad 3.)

 

 53 Des vraies et des fausses idees, 38:185. See my translation, p. 7.

 

 54 Des vraies et des fausses idees, 38:340. See my translation, p. 162. Arnauld does not consider the possibility that our perceptions come from material things because he accepted the Augustinian and Thomistic position that a material thing cannot produce an immaterial effect.

 

 55 On the distinction between causation and causal determination, see G.E.M. Anscombe, ‘Causation and Determination,’ (see n. 30 above).

 

 56 See p. 226 above.

 

 57 Ecrit du pouvoir physique, 10:516.

 

 

4. In What Way Does Efficacious Grace Determine the Will?

 

 

 

One part of Arnauld’s position remains to be explained. In a passage quoted above, Arnauld says that God determines the will by means of efficacious grace.58 Now if this determination is causal, then it seems that Arnauld’s position amounts to this: our meritorious acts of will are free because they are not determined by the nature of their secondary causes, but they are determined by the primary cause. But, then, why is determination by the primary cause any less a threat to free will than determination by secondary causes?

58 See above, n. 52.

 

 59 I would like to thank David Widerker for his help with this paper, thanks to which it is much clearer than it would otherwise have been.

 

 As far as I know, Arnauld does not deal with this question explicitly. But it seems to me that an answer is available to him along the following lines: determination of the will by the primary cause is not, properly speaking, a case of causal determination, but only of logical determination. Let us say that an action, A, of something, x, is logically determined by what another thing, y, does, just in case a proposition describing y’s action entails that x performs A. It should be obvious that if anything causes another thing to act in a given way, the first thing’s causing logically determines the second thing’s action. But logical determination is not the same as causal determination. For example, the proposition

 

that John knows Susan is crossing the street

 

implies

 

that Susan is crossing the street.

 

Hence, if the first proposition is true, John’s knowing logically determines Susan’s crossing the street. But it does not follow that John’s knowing causally determines Susan’s crossing the street.

 

 It is clear that God’s causing an act of will logically determines the act. But what more would be required for God’s causing to causally determine it? Arnauld’s answer, I suggest, is that the act would also have to be determined by the way in which it arises from God, or, if you prefer, by the way in which God works it in us. But Arnauld denies that all our acts of will are determined in that way, when he says, ‘For when God works our volition in us. He brings it about that we will in conformity with the kind of thing we are, that is, by determining ourselves to what we are not determined by nature.’ This, I take it, means that the volition in question is not determined to be the willing of this rather than that – for example, to be the willing of what is good rather than of what is bad – by the way in which it arises from the primary cause. Hence, in the terminology I am now employing, the volition is not causally determined by the primary cause. On this view, when a person performs a free act of will, the possibility that he should will otherwise than he did remains open up to the time he performs the act, and at that time he is not causally determined to will as he does. Of course, on Arnauld’s view, a person does not will anything at all unless he wills, by natural necessity, to be happy. So whenever a person wills anything freely, it is not within his power not to perform any act of will at all, for he must already be willing to be happy. But, on Arnauld’s view, if he does will something freely, then he is not causally determined to will just that.59