Skip to main content Accessibility help
Hostname: page-component-544b6db54f-n9d2k Total loading time: 0.481 Render date: 2021-10-21T15:25:15.368Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "metricsAbstractViews": false, "figures": true, "newCiteModal": false, "newCitedByModal": true, "newEcommerce": true, "newUsageEvents": true }

Chapter 8 - Intellectualist Accounts of the Angelic Fall

from Part III - Angelic Sin

Published online by Cambridge University Press:  26 November 2020

Tobias Hoffmann
The Catholic University of America, Washington DC


Chapters 8–10 constitute Part III, entitled “Angelic Sin,” which raises the issue of how rational agents can do evil under ideal psychological conditions. Chapter 8 is about intellectualist accounts of angelic sin. Since according to these accounts, the will acts as the intellect judges best, evil acts presuppose some cognitive deficiency: either an outright error, or some occurrent nonconsideration that keeps the intellect from making the correct judgment. Thus one difficulty faced by intellectualist thinkers is how the cognitive deficiency can come about – especially since most thinkers here discussed assume that angels are infallible prior to making an evil choice. Another difficulty concerns control of the act. It is assumed that while the angels’ good or evil choice was up to them, the content of their knowledge was not up to them. Aquinas’s solution is that knowledge does not predetermine the use of that knowledge, which is up to the will. By contrast, Godfrey of Fontaines argues that the choice of the angels is caused by the cognized object; he fails to explain, however, how his theory avoids cognitive determinism.

Publisher: Cambridge University Press
Print publication year: 2020

For the study of the moral psychology of free will, explanations of the fall of the angels are no less important than those of the first cause of evil. When it comes to the angels, the question is not only how a good person can do evil for the first time, but also how someone can do so while acting under ideal psychological conditions. Angels are thought to be supremely intelligent; as pure spirits they cannot experience any conflict between rational desire and sensory desire, nor can sensory desire interfere with their practical cognition; as persons whose crucial decision occurred shortly after coming into existence, their decision does not depend on good or bad character formation. In their accounts of angelic sin, medieval thinkers are pressed in a particular way to clarify what motivates rational agents to make evil choices, whether their sin can be a case of clear-eyed evildoing, and how they control their acts.

Certain difficulties in explaining the fall of the angels are particularly challenging for intellectualists, who are considered in this chapter. One difficulty is this:

Infallibility problem: As purely intellectual natures, angels seem unable to experience any cognitive deficiency; so if a sinful act presupposes a cognitive deficiency, then angels cannot sin.

In other words, angels cannot sin because they know better.

Even if their cognition could be deficient, another difficulty is implied in the Socratic deficiency thesis, according to which a volitional deficiency presupposes a cognitive deficiency. But to cause sin – an evil act for which one is responsible – the cognitive deficiency must itself be culpable, for blameless ignorance excuses from sin rather than causing it. This implies a dilemma that is the cognitive counterpart to Augustine’s dilemma:

Blameless ignorance dilemma: Blameless ignorance excuses from sin, and culpable ignorance cannot cause the first sin.

While it is a problem for any person’s first sin, applied to the angels, the problem is this:

Blameless ignorance problem: Angels cannot have any culpable ignorance before their first sin; so if a sinful act presupposes a culpable cognitive deficiency, then angels cannot sin.

In short, angels cannot sin, because even if they do what they should not, they cannot be expected to know better.

A third difficulty is implied in the assumption that the cognized object causes the will’s act. Some medieval theologians assume that the angels obtained all naturally attainable knowledge along with their creation, apart from knowledge of future contingents, which they obtain when the future becomes present. So the increase of knowledge does not depend on the angels, and hence it is not up to them what they know. If angels do not control their knowledge, then they do not control whether a cognized object causes their act. Assuming that sin presupposes the ability to avoid sin, lack of control entails impeccability:

Control problem: Angels cannot know otherwise; so if their actions follow their knowledge, they cannot act otherwise; and since sin presupposes the ability to act otherwise, angels cannot sin.

In short, angels cannot sin, because even if they do what they should not, it was not up to them to know better.

We begin this chapter with Philip the Chancellor, probably the first to formulate the infallibility problem on the authority of Aristotle. We then turn to Albert the Great, whose account resembles Philip’s but is more elaborate. Next, we will consider Aquinas, in whose innovative and resourceful treatment the complexity of angelic sin comes into clear light. Last, we will examine Godfrey of Fontaines, who given his theoretical presuppositions struggles to explain angelic sin, and John of Pouilly, who proposes without any hesitation a strictly intellectualist account of an-gelic sin.

8.1 Philip the Chancellor

Philip the Chancellor introduces into the discussion a popular objection based upon the authority of Aristotle’s dictum that intellectus is always right (cf. An. III.10, 433a26). The word intellectus can mean the power (intellect) or the act (understanding). Philip takes it to mean the power and formulates a version of the infallibility problem: since the angelic appetite is moved only by the intellect (and not, it is implied, by imagination), and since the intellect is always right, angels do not have free decision insofar as it involves “flexibility” toward good or evil (Summa de bono, I: 93 ll. 1–6).

Philip counters that angels must have free decision not only to do good, but also to do evil, for without it, they could not merit. Hence angels can sin. This argument presupposes a symmetrical account of merit, which he professes explicitly further down (I: 201 ll. 35–7): not only does demerit for a bad act require the ability to do good instead, but merit for a good act implies the ability to do evil instead. He points out that unlike angels, humans have two motivational sources, imagination (phantasia, the sensory power of representing an object even in its absence) and practical intelligence (I: 94 ll. 37–45). So, it is implied, humans can sin by following their imagination rather than the judgment of their practical intelligence. Further below, in discussing human sin, Philip explains that imagination presents what is good in the present moment (bonum ut nunc) and intellect presents what is good as such (bonum simpliciter). In support of this distinction he cites the De anima passage more fully: the intellect is always right, while imagination can be right or not (433a26–7). He explains that the intellect can err only by turning to what is mixed with imagination (I: 161 ll. 60–5). When the human will follows the imagination rather than the intellect, it chooses an apparent good, which is not always the true good (I: 166 ll. 48–51). By contrast, although angels do not possess two motivational sources, they have alternative possibilities because they can turn to the unchangeable good (God) or to the changeable good (creatures), in which latter case they sin (I: 94 ll. 45–9, cf. I: 164 ll. 146–8). To the objection concerning the intellect’s infallibility, Philip responds that the intellect is always right not as such, but only as to its “highest part,” to which belongs synderesis (the infallible moral awareness) (I: 94 ll. 50–4).

When Philip is writing, there is not yet a pronounced difference between intellectualist and voluntarist approaches to free will. But Philip’s account of angelic sin has an intellectualist dimension because he recognizes the angels’ infallibility as a threat to their free will and solves the infallibility not by granting that angels act against a correct judgment, but rather by allowing for error in the lower part of the angelic intellect.

8.2 Albert the Great

In Albert the Great, who follows the basic lines of Philip’s solution, the problem of how angelic sin is possible takes on greater importance than in Philip. His first of several discussions of how angelic sin is possible is in De IV coaequaevis, within a Question about the cause of the sin of the angels. A few years later, he pioneers what was to become standard practice: he dedicates a full Question to the possibility of angelic sin, first in his Sentences commentary and then in his lectures on the Divine Names. In his Summa theologiae, whose treatment of angelic sin follows closely that of De IV coaequaevis, he returns once more to this issue within a Question about the cause of angelic sin.Footnote 1 For the idea that the angels have a superior and infallible intellect, Albert uses, like his contemporaries, as an important authority Pseudo-Dionysius’s influential notion that the angelic intellect is godlike (deiformis).Footnote 2 But the formulation of the infallibility problem that he emphasizes is the one based on Aristotle’s authority, which was introduced by Philip. Albert puts it in the Sentences commentary as follows: Aristotle says that the intellect is always right while the imagination is either right or not; in the angels, only the intellect moves the appetite; a sin cannot be elicited by what moves the appetite rightly; hence the angels cannot sin.Footnote 3

In his responses, Albert follows Philip in denying that the intellect is entirely infallible.Footnote 4 His line of thought becomes clearer in light of a passage from De homine, where he asks whether the practical intellect is always right.Footnote 5 In his answer, he clarifies that “practical intellect” can mean three things: first, the rational part of the soul as such; second, that which is in potentiality to all matters of action (just as the theoretical intellect is in potentiality to all things); and third, that which grasps practical principles. In the first sense, as the rational part of the soul as such, the practical intellect is not infallible, as is evinced by the fact that the angelic intellect could turn to evil without being conjoined to imagination. Nor is the practical intellect in the second sense infallible, for matters of action involve mainly particulars, and regarding particulars, one can easily err. Only in the third sense, which according to Albert is the sense intended by Aristotle, is the intellect infallible: it cannot err regarding the universal principles of right (universalia iuris) (De homine, XXVII/2: 483 l. 51 – 484 l. 6). The first practical principles or universal principles of right are, for example, that one must not kill, one must not commit adultery, one must observe justice, one must venerate God, one must honor one’s parents (In De anima III.4.6, VII/1: 235 ll. 30–5).

In his discussions of angelic sin, Albert repeats the idea expressed in the De homine that the intellect is fallible regarding knowledge of particulars. In his Sentences commentary, the explanation of the angelic intellect’s fallibility resembles Philip’s explanation of human sin: the angelic practical intellect is either concerned with the good as such (bonum simpliciter) or with the angels’ own good, which is a “good as to the present moment.” Regarding the good as such, the angelic intellect can only be right, for then it proceeds from the principles of the truth as such and of the good as such (by which Albert seems to intend the first, self-evident, theoretical and practical principles). Regarding the angels’ own good, however, their practical intellect can be deceived (In Sent. II.5.1 sol., p. 111a). In fact, as Albert explains in his De anima commentary, while the “good as such” (bonum simpliciter) is good always and everywhere, this is not the case with the “good as to the present moment.” He cites the famous example that returning borrowed goods to their owner is good, but if the good is a sword and if the owner has gone mad, doing so is bad (In De anima III.4.6, p. 235 ll. 54–75). In the commentary on the Divine Names, Albert explains that deception could happen through mistaken application of universal knowledge to particulars, or by mistakenly relating one particular to another (In DDN IV n. 150 sol. and ad 1, XXXVII/1: 236–7). What here is formulated abstractly corresponds to a hypothesis by Anselm that Albert mentions in De IV coaequaevis: Lucifer knew that his sin deserved punishment, but – applying falsely this knowledge to his personal case – he thought that God would be forbearing (IV.63.3 ad 1 and ad 3, pp. 675b–676b; cf. Anselm, DCD 23).

What is important in all these explanations is that Albert takes the infallibility problem very seriously. He adopts the intellectualist view that the will follows cognition, and hence that deficient willing presupposes deficient cognition. His effort therefore consists in showing that the angelic intellect is not entirely infallible. Nevertheless, in his Sentences commentary, he adds to his standard solution an additional explanation: the angels can sin because their will need not follow their cognition (In Sent. II.5.1 s.c. 2 and ad 4, p. 111a–b). But such a voluntarist explanation is absent from his slightly later account in the commentary on the Divine Names, where he argues that since volitions follow from knowledge, the abandonment of the desire of the good follows from mistaken thinking.Footnote 6

Having offered a solution to the infallibility problem, the next task would be to explain how the angels could have avoided the error that led to their sin, and whether failure to avoid it is itself a sin. But Albert does not provide this explanation, and so he leaves the blameless ignorance problem unaddressed.

8.3 Thomas Aquinas

In treating the various problems implied in the narrative of angelic sin, Aquinas raises the discussion to a higher philosophical level than his immediate predecessors. Compared to them, Aquinas’s theory of angelic psychology poses a greater challenge to explaining angelic sin, and he makes greater use of Aristotle’s action theory. By contrast, he pays almost no attention to Anselm’s De casu diaboli, which was, however, an important source for his contemporaries. We begin with Aquinas’s account of angelic psychology, which sets narrow boundaries for his solutions to the philosophical problems implied in angelic sin.Footnote 7

Angelic Psychology

Aquinas brings out in greater detail than his predecessors the optimal psychological makeup of angels. He holds, of course, that no angel is evil by nature.Footnote 8 In addition, he argues that the angels cannot have any natural inclination to evil. As he explains, every nature is good, and so nothing can have a natural inclination to evil, except incidentally, insofar as it is inclined to a particular good that happens to be contrary to some other good. Only things that are composed of two natures can have such conflicting goods. In human beings, sensory pleasure, a particular good, can be contrary to the good of reason, which is the good simply speaking (bonum simpliciter); thus in humans, there is a natural inclination to the sensory good that can be contrary to the good of reason (or of the intellect). But as purely intellectual beings, angels can only be inclined to the good of the intellect, and thus to the good simply speaking. Hence they cannot be naturally inclined to evil.Footnote 9

Aquinas thinks that angels cannot desire evil under the aspect of the good, for they cannot have erroneous beliefs. Angelic infallibility follows from certain assumptions of Aquinas’s theory of angelic knowledge. He holds that angels, like human beings, know by means of “intelligible species” that inform their intellect, that is, by means of likenesses of essences. Unlike humans, angels do not acquire intelligible species over time, but possess all of them from their creation (ST I.55.1–3). Based in part on the authority of Pseudo-Dionysius (DDN VII.2), Aquinas holds that angels know nondiscursively, that is, they do not gain knowledge of things unknown through things already known, as humans do through syllogistic reasoning (ST I.58.3). Angels cannot err, for they cannot falsely affirm or deny a predicate of a subject, because they know simultaneously and nondiscursively all that can be affirmed or denied of the essences they know through the intelligible species (ST I.58.4–5) – except for future contingents (DM 16.7 ad 6). Thus they have perfect knowledge of everything that falls within the natural scope of their intellect (DM 16.6c., p. 310 ll. 233–44). Nevertheless, the knowledge of higher angels extends further than that of lower angels, because they possess more universal intelligible species (ST I.55.3; DM 16.4c., p. 299 ll. 319–30). Yet no angel is omniscient. Although they know by nature that God exists,Footnote 10 at first, they did not enjoy the beatific vision (ST I.62.1), and so they did not know any divine mysteries unless they were revealed to them (ST I.57.5). Only if they presumptuously judge about things that fall outside of the scope of their natural knowledge can they err. For example, they can believe that a dead man (that is, Christ) will not rise from the dead, or that Christ is not God. Such presumptuous judgments cannot come about without an evil will (ST I.58.5c.; DM 16.6c., p. 310 ll. 255–91).

Despite the angelic intellect’s perfection, Aquinas thinks that it is limited in that it cannot simultaneously consider everything it knows habitually, just as we cannot simultaneously think about all we know. Aquinas professes a theory of undivided thinking: what an angel is actually thinking about at a given moment is restricted to what is knowable through a single intelligible species (In Sent. II.3.3.4; ST I.58.2).Footnote 11 As he explains, this restriction is due to the fact that the intelligible species is the “form” of the intellect, that is, its cognitive configuration, and nothing can be per-fected by several forms, just as a body cannot be configured by different configurations. Thus, angels can think of different things at a time, but only as many as are knowable through a single species – which, however, is no small number.Footnote 12 Their habitual knowledge extends to everything that is naturally knowable to them; their occurrent consideration does not (DM 16.6 ad 4).

Given the optimal conditions of the angelic intellect and will, it is difficult to see how angels could sin. Philip and Albert had made concessions to angelic infallibility prior to sin; Aquinas does not. He addresses the possibility of angelic sin both in general, arguing that all rational creatures are in principle able to sin, and in particular, showing how the angels could sin despite their infallibility concerning naturally knowable truths.

The Peccability of Rational Creatures

Augustine traced the rational creatures’ ability to sin to their being created from nothing (ciu. XII.6). Aquinas adds to this explanation that for rational creatures being created implies being subject to God’s rule and measure. Nonhuman animals and God are not under a higher rule, and hence they cannot have an evil appetite. The good of an animal is to follow its sensory appetites, whereas God is his own rule and hence cannot sin, just as a person cutting wood could not fail if his or her own hand were the rule for correct cutting.Footnote 13 Aquinas consistently applies his theory of the natural peccability of rational creatures to the case of angels.Footnote 14

Aquinas’s Early Treatment of the Infallibility Problem

Aquinas discusses the problem of how angelic sin is possible repeatedly throughout his career.Footnote 15 By examining his early treatment in the Sentences commentary and his late treatment in the Prima pars of the Summa theologiae and the De malo, we can observe the development of his theory and see how it relates to discussions before and after him.

The basic narrative Aquinas presents in the Sentences commentary remains substantially the same in later treatments, whereas his thought about how angels could sin undergoes some development. Aquinas holds that all angels are in a state of happiness from their creation, which, however, is not the highest happiness of which they are capable (In Sent. II.4.1.1). What Lucifer, the highest angel (In Sent. II.6.1.1), wrongly desired was his preeminence in happiness and goodness compared to all lower creatures, and he wanted to have his ultimate perfection, complete happiness, by his natural means alone rather than through grace. So he desired in a way to be like God, for to have complete happiness by nature rather than by grace is unique to God (In Sent. II.5.1.2c.).Footnote 16 Lucifer’s disordered desire for his excellence, triggered by the consideration of his own beauty, is pride (In Sent. II.5.1.3c.). Lucifer also wanted to be the cause of the ultimate perfection of the lower angels, and when he expressed this desire to them, some consented (In Sent. II.6.1.2).

Regarding the problem of how angels can sin, Aquinas formulates an opening argument that repeats the objection discussed by Philip and Albert: angels cannot sin because their appetite follows their intellectual cognition, which cannot be false (In Sent. II.5.1.1 arg. 4). In the solution, Aquinas states that “there cannot be sin in the will, unless there is somehow deception in reason, hence ‘every evildoer is somehow ignorant’,” quoting the famous Aristotelian adage (EN III.1, 1110b28–9). His solution builds upon Aristotle’s explanation of how incontinence is possible. Recall that Aristotle distinguished between merely having knowledge and having and exercising knowledge, and furthermore between knowing universal propositions (dry food is good) and knowing particular propositions (this is dry food). According to Aristotle, because of passion, the incontinent do not apply their habitual knowledge to the case at hand (see Chapter 1, pp. 256). Aquinas mentions the two distinctions and puts the brunt of the weight on the first one: someone can have habitually (in habitu) the correct assessment, even concerning a particular matter of action, but be impeded from actually considering it, because passion “fetters” the intellect. Angels lack passions, but their intellect can nevertheless be fettered. Based on his theory of undivided thinking, Aquinas argues that by considering one thing, an angel is drawn away from considering other things. The angels can sin by considering an aspect according to which something is choiceworthy, although it is not to be chosen, all things considered. Thus they commit an “error in choosing” (error electionis) (In Sent. II.5.1.1c.). In response to the opening argument, Aquinas clarifies how such nonconsideration can lead to a deficiency in the intellect without causing it to err: a doctor can fail by correctly judging something to be beneficial to a sick person insofar as a certain sickness is concerned, which, however, is not beneficial to the person in view of another sickness the doctor does not consider. Just so, the angelic intellect can be deficient by failing to consider something, although it cannot err by judging something to be the case that is not the case (In Sent. II.5.1.1 ad 4). An “error in choosing” need not be caused by a false intellectual judgment, but rather may be caused simply by the failure to consider properly what was to be chosen (In Sent. II.39.3.2 ad 5).

Aquinas’s Mature Treatment of the Infallibility Problem

Aquinas’s mature account of the fall, in the Summa theologiae and the De malo, uses again the idea that the angels sinned not because of a mistaken judgment but rather because of some lack of consideration. Aquinas now clarifies what it was that the angels overlooked: the divine rule by which their will was to be regulated. In the Summa theologiae, he gives this explanation in response to an opening argument that formulates a major difficulty concerning angelic sin: the angels could not sin by wanting evil as such, for evil under the aspect of evil is beyond the will’s scope; nor could they sin by wanting evil under the aspect of the good, because mistaking an evil for a good is impossible for their intellect, which – at least before their first sin – could not err (ST I.63.1 arg. 4). So they could only sin by wanting a true good. In the response, the idea of nonconsideration of the rule has a double function: concerning moral theory, it explains why the desire for something truly good can be bad, and concerning moral psychology, it responds to the requirement of some cognitive deficiency for an evil choice.

Sin in an act of free decision can happen in two ways. In one way in that something evil is chosen, as when a man sins in choosing adultery, which as such is evil. And such sin proceeds always from some ignorance or error; otherwise that which is evil would not be chosen as a good.... In another way it happens that one sins by free decision by choosing something good in itself, but not according to a proper measure or rule. In this case, the defect that induces sin is only on the part of the choice that is not properly regulated, not on the part of the thing chosen; as if one were to pray, but without heeding the order established by the Church. Such a sin does not presuppose ignorance, but merely lack of consideration of the things that ought to be considered. This is the way the angel sinned: by seeking his own good, by his own free decision, without ordering [the act] according to the rule of the divine will.

(ST I.63.1 ad 4)

As in his earlier discussions, Aquinas holds that the good the sinning angels desired was complete happiness, which they intended to attain by their natural power. According to the divine rule, however, complete happiness was to be obtained as a gift freely bestowed on them by God’s grace.Footnote 17 Aquinas identifies the nonobservance of the higher rule with the sin of pride (ST I.63.2c.).

For our purposes, what is most important is the psychological function of nonconsideration of the rule in angelic sin. In the above quote, Aquinas draws the parallel between an adulterer’s ignorance and the evil angels’ nonconsideration, implying that the Socratic deficiency thesis applies to both cases. He suggests that, just as the adulterer would not have sinned if he or she had known better, so also the angels would not have sinned if they had attended to the divine rule. In the De malo, Aquinas repeats essentially the same solution. The angels failed to cognize the rule and what it ordered, which is a presupposition for sin because “in a sin, the deficiency of the intellect or reason and of the will are always proportionally tied in with another.” Thus sin followed in the angels’ will upon insufficient consideration of the order of the divine governance (DM 16.2 ad 4 and ad 5).

Why could the angels not have sinned in full consideration of the rule, contemptuously dismissing it? Aquinas does not directly answer this question, but he provides some hints. By sinning, Lucifer “withdrew from the order of divine justice, and so he abandoned what was better [deseruit meliora], for the rule of divine justice is better than the rule of the angelic will” (DM 16.3 ad 10). Hence, we may assume, if Lucifer had considered the rule, he could have recognized that its ordinance was better than his own plan. Perhaps by considering the rule, he could have concluded – by understanding or by faith – that by observing it, he only had to gain. In fact, according to Aquinas, had Lucifer not sinned, he would have attained precisely what he inordinately desired; Aquinas supports this idea with an oft-cited claim by Anselm of Canterbury: “he desired what he would have obtained if he had persevered [stetisset]” (ST I.63.3c.; cf. Anselm, DCD 6, I: 243–4). If Lucifer had recognized what was in his better interest, he would have acted accordingly. Aquinas thinks that not even the already fallen angels act intentionally against their own interest (perverse as it may be); for example, if they had realized the effects of Christ’s crucifixion for human salvation (which they dread), they would never have wanted his crucifixion (ST I.64.1 ad 4).

That the angels could only sin while not considering the divine rule does not mean, however, that they sinned inadvertently. In Aquinas’s account, the evil angels were fully aware that they committed a “mortal sin,” that is, one that turns them away from God. Human beings can sin “venially,” that is, without turning away from God, because they do not always realize how a particular act affects their relationship with God. In human beings, such venial sins can happen even without their awareness. But angels, who do not think discursively, cannot know something without knowing all its implications; thus, when they fail to refer a desired good to God, they are aware that this failure implies turning away from God (DM 7.9c. and ad 2).

Aquinas’s account of nonconsideration in angelic sin can be compared to people who damage a new technological device because they did not read the instruction manual. In the normal case, this happens not because they read the manual and intentionally disregard its instructions, nor because they are unaware of the existence of the manual, but rather because they feel confident that they can use the device without further instructions. The question, then, is what one should be blamed for: for having disregarded the manual, for the improper use of the device that follows it, or for both. Regarding the angels, the question is whether the nonconsideration of the rule is itself a sin (of negligent omission), or whether the sin is only the unregulated act, or both the act and the nonconsideration that made it possible. The issue here is the blameless ignorance dilemma: blameless nonconsideration of the rule excuses from sin; culpable nonconsideration cannot be a presupposition for the first sin. As already mentioned, this difficulty is analogous to Augustine’s dilemma: a good will cannot cause an evil will, and an evil will cannot cause the first evil will. Aquinas’s solution to Augustine’s dilemma (see Chapter 7, pp. 1847) applies in a similar way to the blameless ignorance dilemma. Nonconsideration as such is no sin; it becomes sinful only when one does an act that is to be regulated. But even then, it is not a stand-alone sin (a “special sin,” in medieval jargon), but only a dimension of the actual sin. Aquinas does not spell out this solution in his late treatments of angelic sin, but he did formulate it in the Sentences commentary. In an opening argument it is maintained that Lucifer’s first sin consisted in the sin of omission rather than in pride, for he failed to consult a divine ordinance that would have forbidden the desiring of something beyond what was ordered. In response, Aquinas clarifies that omission is in one sense a special sin and in another sense an aspect of every sin. As a special sin, omission concerns something that was specifically enjoined; as an aspect of every sin, omission is the failure to observe a relevant circumstance of the act. The latter case applies to Lucifer’s sin.Footnote 18

The Control Problem

The idea of nonconsideration of the rule is so important in Aquinas because he assumes a tight connection between cognition and choice. Nevertheless, the choice of the angels cannot be fully traced to their knowledge. I will first give a thumbnail sketch and then a fuller account of why this is so.

Whereas the angels controlled their choice (as Aquinas assumes), they did not control their knowledge. But they controlled the use of their knowledge, and so they controlled how habitual knowledge becomes an occurrent practical judgment that is followed by the corresponding choice. We have already seen an instance in which Aquinas traces the use of knowledge to the will and no further: in De malo 1.3 he writes that the use or nonuse of the moral rule depends on the will’s freedom, and no further cause need be searched (see Chapter 7, p. 186). The case of angelic sin shows even more forcefully that there is no room for further tracing of the first choice beyond the will. In fact, before their first choice, there is no distinction between good and evil angels (In Sent. II.4.1.2c.). According to Aquinas, after creation and before their first choice, there is only one moment in the angelic existence, in which all angels were good. Immediately thereafter, in the second moment of their existence, all angels made either a good or evil choice (ST I.63.6; cf. DM 16.4c., p. 300 ll. 349–74). Thus, Aquinas locates an angel’s control of his choice in the way his will uses his knowledge.

Now we must see this in greater detail. Aquinas is committed to the view that the angels could do otherwise. That the angels controlled their choice is implied in the fact that the evil choice was a sin. For Aquinas, what distinguishes moral fault from mere mishap is that the act in question is imputed to the agent. Such imputation presupposes that one control the act; as he phrases it, the act must be in one’s power (in potestate), that is, one must have mastery (dominium) over one’s act.Footnote 19 Aquinas expresses the angels’ control by claiming that they sinned by their own free decision, by which they could choose to sin or not to sin.Footnote 20

That the angels did not control their habitual knowledge follows from the fact that they received all intelligible species belonging to their natural knowledge from their first moment of existence. This knowledge is continually updated as future contingents become actual in the present moment (DM 16.7 ad 6). Other than through such updates, angelic knowledge can increase only through divine revelation. Neither change in knowledge depends on them (except for knowledge of their own choices and foreseeable consequences of their choices). Hence, prior to making their first choice, they did not control their habitual knowledge. But ex hypothesi, they controlled this choice. Their situation is roughly analogous to God’s, who cannot know otherwise, but who can nevertheless will otherwise. Aquinas writes: “Whatever God knows, he knows by necessity; however, he does not by necessity will whatever he wills” (ST I.19.3 ad 6).

So when the angels make their free choice, there is a transition from necessity (of knowing) to contingency (of willing). Based on Aquinas’s account of angelic sin and on what he says elsewhere about cognition and free decision, this transition happens with the “use” of their knowledge or the “exercise” of their intellect, through which they actually consider something they know habitually. In his mature account, Aquinas explains that the good and evil angels differed in whether they considered the rule governing their choice (e.g., ST I.63.1 ad 4); earlier, he argued that they differed in whether they considered all the conditions relevant to making a good choice (In Sent. II.5.1.1). So depending on how the angels use their knowledge, they judge what is worth choosing and act accordingly.Footnote 21

Aquinas traces the way in which rational creatures use their knowledge to the will. Knowing something habitually (scire) means possessing its intelligible species without actually considering it.Footnote 22 But thinking about something (intelligere, cogitare) consists in using an intelligible species, and this use depends on the will.Footnote 23 How the will uses an intelligible species – whether or not it uses a certain species, and what precisely it considers by means of this species – is not predetermined by that intelligible species or by any other intelligible species. That Aquinas holds this view is implied in a consideration he makes about whether someone who has access to another person’s intelligible species – in the case he considers, an angel knowing a species in a human mind – would know what that person is thinking about. Aquinas denies this, precisely because by knowing someone’s species one does not yet know how that person’s will uses the species (DM 16.8c., p. 321 ll. 200–41).Footnote 24

Another doctrine of Aquinas makes clear that he does not think that what a person wills follows exclusively from what the person knows. As we have seen, in his explanation of how the will is moved (that is, actualized) by the intellect and how it is moved by itself, Aquinas writes that the will moves itself to the exercise of the act, albeit by means of making the intellect deliberate about whether to exercise its act (Chapter 2, p. 50). To avoid the regress implied if every exercise of a power required a previous deliberation, including the exercise of the activity of deliberation, Aquinas traces the will’s exercise of its act ultimately to God moving it. While tracing it to God bears its own theoretical difficulties, what is important is that Aquinas does not trace the exercise of the will’s act exclusively to cognition (DM 6c.; ST I-II 9.6).

It appears, then, that according to Aquinas the angels’ choice at the second moment of their existence was not predetermined by their knowledge in the first moment of their existence. Nor does any other fact in the first moment of their existence determine their choice in the second moment. For this reason, Aquinas argues that prior to their fateful choice, the angels themselves could not know – not even conjecturally – what they would choose; hence they could not foresee whether they would fall or persevere in the good (In Sent. II.4.1.2c.). Aquinas’s account is intellectualist in its general outlook, insofar as he does not admit that angels sin by willing against an all-things-considered judgment of what is worth willing. But his account has a voluntarist dimension in that he traces the angels’ use of their habitual knowledge, especially of the divine rule that was accessible to them but not necessarily considered, to their will and no further. Thus, the control the angels had of their choice and of its antecedents is primitive.

Yet the evil angels did not act without a motive. Lucifer sinned because he took pride in his superiority over others (ST I.63.7c., In Sent. II.5.1.2c.). In general, the evil angels were lured to sin by their own beauty (DM 16.2 ad 13, cf. In Sent. II.5.1.3c.). But they were not drawn by these motives necessarily; indeed, the good angels were beautiful, too, yet their beauty did not cause them to sin. While the motives explain the choice, there is no explanation for why an angel acts on this motive rather than on that. Nor is there a contrastive explanation for why Lucifer sinned rather than not, or why Lucifer sinned but not Michael. Their choice can be made intelligible only by explaining how they could act in one way or the other, but not why they did act in the way they did.

8.4 Godfrey of Fontaines

We will now see whether Godfrey of Fontaines and John of Pouilly manage to explain angelic sin without making any voluntarist concessions.

The only discussion of angelic sin by Godfrey of Fontaines that has come down to us is a brief reply to an opening argument in his Quodlibet VI, Question 7. This Question, as we have seen in Section 4.3, contains a lengthy discussion of the impossibility of per se self-motion. Against Godfrey’s idea that self-motion is possible only where the mover and the moved are in distinct subjects, the opponent refers to angelic cognition and angelic sin as counterexamples: an angel who because of an intelligible species understands something in potentiality moves himself to understand the thing in actuality; an evil angel produces in himself an evil volition (Quodl. VI.7 arg. 1, PhB III: 149).Footnote 25

The issue implied in this argument is the control problem: if the angels are to be blamed for their sin, then they cannot have undergone evil, but it must have been up to them to have an evil will or not. But since Godfrey denies that mover and moved can be in the same subject, he rejects the idea that the will of the evil angels caused their volition.

Godfrey starts his reply to the opening argument with the remark that the philosophers denied any transition from potentiality to actuality in separate substances (that is, in immaterial substances: God and angels). Moral deficiency in separate substances, it is implied, is unthinkable to them. But unlike them, Godfrey takes as his guide not natural reason but Sacred Scripture, and hence he accepts the fall of the angels on the authority of faith. He does not deny the angels’ transition from potentiality to actuality; he only denies that they moved themselves from potentiality to actuality (Quodl. VI.7 ad 1, PhB III: 168). His answer builds upon Aqui-nas’s theory:

I say that their fall does not prove that the same subject moves itself, because the devil sinned only by his spontaneous will, desiring something in a manner in which he was not supposed to desire it. Under the presupposition of the defectibility of the will and nonconsideration and some sort of negligence by reason, the cognized object caused in him the act of willing. By this act he sinned, because he desired this object in a manner he was not supposed to – not because of an error or ignorance that would be temporally antecedent, but rather from a simple unawareness [nescientia] due to a deficiency of considering the manner [of proper desire] he was supposed to consider. But upon this unawareness or lack of consideration there followed an error or ignorance in the moment he consented. Although this error or ignorance was temporally simultaneous with the consent, according to nature it was prior [to consent].


Even concerning angelic sin, Godfrey maintains his conviction that a volition is caused by the cognized object – by efficient causation, as is implied and as he affirms explicitly earlier in the same Question, although not in reference to the angels’ volitions. He does not explain, however, how the angels can control what they consider and which object causes their volition; his prohibition of self-motion impedes such an explanation. In his further discussion, he insists that just as the will does not move itself to will something, so too the intellect does not move itself to understand something. Instead, it is the object that moves the intellect to its cognition and the will to its volition. As we have seen in Section 4.3, Godfrey is adamant that the principle that something cannot reduce itself from potentiality to actuality applies to all of being, and thus also to angels and the human soul. He also insists that first and most certain metaphysical principles must not be denied, even if they make it difficult to explain how to safeguard free will (Quodl. VI.7 ad 1, PhB III: 170; cf. Chapter 4, pp. 1089).

In the above block quote Godfrey furthermore affirms that nonconsideration led the angels into error. While Aquinas excluded fallibility from angels prior to sinning, Godfrey does attribute error to them, perhaps because he takes judgment-volition conformity to imply that the devil’s perverse volition presupposes the mistaken practical judgment that it is good to will in this manner. His disciple John of Pouilly will make this point explicitly.

Godfrey’s discussion of angelic sin is brief but important, because it brings out the tension between the legitimate reservations against admitting per se self-motion and the attempt to explain a free choice among alternatives. What is significant in Godfrey’s solution is above all what it does not explain: how the angels control which choice they make. His position is vulnerable to the objection that it fails to avoid intellectual determinism.

8.5 John of Pouilly

Whereas Godfrey discusses angelic sin because he was challenged, John of Pouilly takes it up of his own accord, to reject the claim that angelic sin and the sin of the progenitors Adam and Eve are a counterproof to judgment-volition conformity.

Pouilly develops his position in response to Gonsalvus of Spain. As part of a series of arguments to show that the will can act against the judgment of reason, Gonsalvus maintains that, given the Augustinian deficiency thesis that deficient cognition presupposes deficient willing, the angels and progenitors suffered no ignorance before their sin, and so they could only have sinned by acting against their better knowledge (QD q. 8c., ed. Amorós, p. 128). Gonsalvus’s argument seems inspired by Henry of Ghent (see Chapter 3, pp. 6970); it was in general popular among voluntarists, as we will see in Chapter 9.

Against Gonsalvus’s argument, Pouilly defends the Socratic deficiency thesis in order to safeguard judgment-volition conformity. Pouilly argues that one cannot will what is unknown, and hence one cannot will something under an aspect under which it is not known. In positive terms, what one wills must be cognized and judged as to be willed under the very aspect under which one wills it. The angels and progenitors, then, no matter how they sinned, could not have sinned without making a cognitive error. If they desired an evil as an apparent good, they erred. If they desired a good, but not in the way in which it was meant to be desired, then they erred as well. Assuming that they desired their own perfection as a good in itself rather than as subordinate to a higher good, they erred because they judged that this particular good was to be desired for its own sake, when in fact it was to be desired for the sake of something else. According to Pouilly, Sacred Scripture confirms that the first sin was caused by error. When God asks Eve why she took of the forbidden fruit, she answered: “because the serpent deceived me,” rather than “because I had a bad will” (cf. Genesis 3: 13).Footnote 26 Pouilly thus sides with Godfrey that sin presupposes an outright error, not merely some nonconsideration.

Pouilly knows that in attributing fallibility to the angels and progenitors, he goes against the majority view. He therefore offers his own exegesis of the proof text from Augustine on which some theologians base the Augustinian deficiency thesis. The text is usually quoted as affirming that fallibility in Adam is punishment for his sin; from this it is concluded that before sinning, Adam was infallible; hence error presupposes an evil will (cf. Chapter 3, pp. 6970). Pouilly observes that so stated, the quotation is incomplete. Augustine does not write that the punishment for sin is to err, simply speaking, but rather to err unwillingly (invitus) (see indeed lib. arb. III.18.52.179). It was possible to err before sin; but the point is that then it was also possible not to err, and error was voluntary. Now, in contrast to Adam, we err willy-nilly (Quodl. II.11 concl. 8, A 112vb, P 55va–vb). Thus, for Pouilly, the infallibility problem is a false problem. One need not explain how the angels could sin despite infallibility, for they were not in-fallible in the first place. In making these points, Pouilly does not con-sider other arguments for angelic infallibility, such as those of Aquinas, that are based upon the working of the angelic intellect.

If the error that led to the first sin was voluntary, how could the angels and progenitors have guarded against it? Pouilly addresses this question in another context: while discussing the cause of an evil will, he comments that his theory on that subject applies also to the sin of angels and progenitors (see Chapter 7, p. 193). The immediate cause of an evil will is error or deception, the cause of which is negligence. We control whether we are negligent or diligent through intellect and will, by which we can apply good objects and disapply bad objects. By deliberating diligently we can avoid making an error in reason and then committing a subsequent sin in the will (Quodl. IV.6c., P 120rb–va, N 91rb–va).Footnote 27 Presumably, the case of the angels is analogous for Pouilly: by being diligent, they could apply an object in conjunction with the fitting order in which it was to be desired, or through negligence, they could apply an object that omits the correct order. As we have seen in Section 7.9, Pouilly holds that negligence is not itself to be traced to a per se cause.

It seems, then, that for Pouilly, the control of being either negligent or diligent is primitive, and likewise the control of applying or disapplying an object to the will. If this is so, he is not so far from voluntarist thinkers, except that Pouilly traces control equally to intellect and will, whereas voluntarists trace it mainly to the will.

* * *

Of the three problems mentioned at the beginning of the chapter, Philip, Albert, Aquinas, and Pouilly address above all the infallibility problem. Philip and Albert conceded that angels can err about particulars and hence about the cognitive antecedents of a choice. Aquinas did not allow for angelic fallibility, only for nonconsideration of habitual knowledge. Pouilly’s concern is less to explain angelic sin than to show that it does not demonstrate the will’s ability to act against the judgment of reason, and he is willing to sacrifice the angels’ infallibility for that.

Aquinas does not explicitly address the control problem, but his moral psychology makes room for angels controlling their choices by controlling what they actually consider. The control problem poses a serious difficulty for Godfrey, who all but admits that his premises leave him unable to explain how an angel can sin of his own free will. Pouilly feels more confident than his master about the problem of the angels’ (and progenitors’) sin. But he can safeguard their free decision only by covertly making the power to apply or disapply an object to the will primitive.

Control of choices among alternatives comes with a price: although a freely made choice is made for a reason, it remains at bottom unexplained why the agent acted on this rather than on that reason. Intellectualists are in general more reluctant to accept that free choices escape full intelligibility. In the case of the angels, however, who under the same initial conditions made opposite choices, this acknowledgement is inevitable. Voluntarist thinkers are generally more forthright in acknowledging the ultimate inexplicability of acts of the will in general and of evil acts in particular. We now turn to their accounts of how a sinful choice can originate in the angelic mind.


1 De IV coaequaevis exists in two redactions. The first, which is unedited, was written before 1241; the second, on which my exposition is based, is contemporaneous to In Sent. II (1246/47); see Reference RistRigo 2005, 347–58. For the dating of In DDN (1250) and of Summa theologiae (after 1268), see Footnote Chapter 7, notes 13 and Footnote 18, respectively.

2 De IV coaequaevis IV.63.3 arg. 2, Borgnet XXXIV: 674b; In Sent. II.5.1 arg. 2, Borgnet XXVII: 110b; ST II.5.21.3 arg. 2, Borgnet XXXII: 262b–263a; cf. Pseudo-Dionysius, DDN VII.2, 869C.

3 In Sent. II.5.1 arg. 1, Borgnet XXVII: 110b; see also De IV coaequaevis IV.63.3 arg. 1, Borgnet XXXIV: 674b; In DDN IV n. 149 arg. 1, XXXVII/1: 236a; ST II.5.21.3 arg. 1, Borgnet XXXII: 262b.

4 De IV coaequaevis IV.63.3 ad 1, p. 675b; In Sent. II.5.1 sol., p. 111a; In DDN IV n. 150 ad 1, p. 237a; ST II.5.21.3 ad 1, pp. 263b–264a.

5 De homine was written after De IV coaequaevis (first redaction) and before the Sentences commentary; cf. Footnote Chapter 2, note 10.

6 In DDN IV n. 149 arg. in opp., p. 236 ll. 68–74, endorsed on p. 237 l. 49.

7 On Aquinas’s account of angelic sin in general, see Reference MartelMaritain 1959, Reference BoninoBonino 2016 ch. 10 and the literature he indicates on p. 197, Footnote note 25, and Reference D’ErcoleD’Ercole 2017, 75–104; for the control problem in Aquinas’s account of angelic sin, see Reference HoffmannHoffmann 2007 and Reference PerkamsPeck 2014.

8 SCG III.107; In Ioannem 8.6 n. 1246; In DDN IV.19 nn. 529 and 535; ST I.63.4; DM 16.2c., XXIII: 288 ll. 191–9; DSS 17, XL: D70–1; DSS 20, pp. D76–7 ll. 27–64.

9 DM 16.2c., p. 288 ll. 200–57; DSS 20, p. D77 ll. 65–94; see also ST I.63.4.

10 ST I.56.3; DM 16.3c., p. 293 ll. 174–80; DM 16.4 ad 14.

11 Reference DubouclezDubouclez 2014 gives a lucid explanation of Aquinas’s theory of undivided thinking and documents that it generated controversy through the fourteenth to seventeenth centuries and became widely rejected.

12 ST I.58.4c. and ad 1; DM 16.4c., p. 299 ll. 335–9. On Aquinas’s theory of angelic knowledge, see for more details Reference SzlachtaSuarez-Nani 2002b, 17–75, and Reference GorisGoris 2012.

13 DM 1.3 ad 9; DM 16.2c., p. 289 ll. 277–87. Aquinas here implicitly rejects Aristotle’s theory that the virtuous person embodies the “rule and measure” of what is truly good (EN III.4, 1113a28–33). For earlier, different explanations, see In Sent. II.23.1.1 and DV 24.7.

14 SCG III.109; ST I.63.1c.; DM 16.2c., p. 289 ll. 300–5. Aquinas’s theory of the peccability of angels generated a lively debate in the mid-twentieth century because of its implications for the relation between nature and grace. The issue is whether angels are capable of sinning only if they are destined by grace toward a supernatural final end. See Reference MaritainMarieb 1964 and the literature he cites.

15 In Sent. II.5.1.1; SCG III.108–10; ST I.63.1; DM 16.2; DSS 20 (unfinished).

16 Regarding Aquinas’s theory of the object of Lucifer’s desire, see Reference PironPini 2013, 66–71.

17 ST I.63.3; DM 16.2 ad 4, p. 289 ll. 358–61; DM 16.3c., p. 294 ll. 219–29 and ad 1.

18 In Sent. II.5.1.3 arg. 4 and ad 4. Cf. Footnote Chapter 7, note 28.

19 ST I.48.5c.; ST I-II.21.2c.; DM 2.2c., p. 33 ll. 136–42.

20 In Sent. II.5.1.1 s.c. 2 and ad 1; In DDN IV.19 n. 541; ST I.63.1 ad 2 and ad 4; ST I.63.7c. and ad 3; ST I.64.2c.; DM 16.5c., pp. 304–5 ll. 215–52. Cf. DM 16.4c., p. 300 ll. 364–5. For the relation between moral responsibility and free decision in Aquinas, see Footnote Chapter 2, note 26.

21 In another context, Aquinas describes how the transition from habitual knowledge to a particular choice is mediated by a practical judgment that depends on the use one makes of one’s knowledge; see DM 3.9 ad 7.

22 DV 10.2c., XXII: 301–2 ll. 175–81 and ad 4; ST I.79.6c. and ad 3.

23 ST I.57.4c.; ST I-II.57.1c.; DM 16.8c., p. 321 ll. 193–9, 215–21.

24 For an excellent discussion of Aquinas’s theory of mind-reading and its implications for his account of cognition, see Reference CoryCory 2016. See also ST I.107.1, concerning angels reading other angels’ minds.

25 Duns Scotus will formulate a similar objection against Godfrey; see Lect. II.25 n. 40, XIX: 241; Rep. II-A.25 n. 9, Viv. XXIII: 121a–b; Ord. I.3.3.2 nn. 430–2, III: 262–3.

26 Quodl. II.11 concl. 8, A 112vb, P 55va. This passage is edited in Reference Hoffmann and TregoHoffmann forthcoming b, Appendix B.

27 The cited passage is edited in Hoffmann Reference Hoffmann and Tregoforthcoming b, Appendix C.

You have Access

Send book to Kindle

To send this book to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle.

Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats

Send book to Dropbox

To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Dropbox.

Available formats

Send book to Google Drive

To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Google Drive.

Available formats