The Indetermination of Reason and the Role
of the Will in Aquinas's Account
of Human Freedom



Thomas Aquinas argues that human choices are made by the will and reason working together. It is easy to misinterpret his argument and suppose that the reason alone works out what should be done while the will simply ratifies this. Instead Aquinas believes that in practical matters the reason is often undetermined since it arrives at many simultaneous conclusions. This is the often unacknowledged heart of Aquinas's account of freedom. All these simultaneous rational conclusions derive from the objective circumstances of the world; each one could give rise to a different rationally justified course of action; yet only one can be acted upon. The reason cannot decide between them. It is the will that accepts and affirms one of these conclusions and gives force to the reasonableness of one course of action. This is why a choice is always rational and personally willed – which is what makes it free. The indetermination of reason is what allows the future to be open-ended for the deliberating agent; it allows past and present to be interpreted in different ways, each of which has its own coherence and rationality. In this way Aquinas's account of human freedom avoids both an irrational voluntarism and a deterministic intellectualism

1. Introduction

Thomas Aquinas argues that human choices are made by the will and reason working together. It is easy to misinterpret his argument and suppose that the reason alone works out what should be done while the will simply ratifies this. Instead Aquinas believes that in practical matters the reason is often undetermined since it arrives at many simultaneous conclusions. This is the often unacknowledged heart of Aquinas's account of freedom. All these simultaneous rational conclusions derive from the objective circumstances of the world; each one could give rise to a different rationally justified course of action; yet only one can be acted upon – and sometimes the reason cannot decide between them. It is the will that accepts and affirms one of these conclusions and gives force to the reasonableness of one course of action. This is why a choice is always rational and personally willed – which is what makes it free.

The indetermination of reason is what allows the future to be open-ended for the deliberating agent; it allows past and present to be interpreted in different ways, each of which has its own coherence and rationality. In this way Aquinas's account of human freedom avoids both an irrational voluntarism and a deterministic intellectualism. Human beings seek (through the will) a good (understood by intellect and reason) – the two elements of choice are inseparable, yet they are also distinct. This article will defend this interpretation of Aquinas's account of freedom and draw out some of its implications by examining some of his key texts.1

A brief introductory note about language is required here. Contemporary English-speaking philosophy tends to discuss questions of human freedom and action under the heading ‘freedom of the will’. Aquinas inherits a tradition which refers to similar issues under the heading liberum arbitrium, which is preserved in the contemporary French le libre arbitre.2 The Latin does not contain the word ‘will’, voluntas. It was a matter of debate whether the will was free, or the reason, or some other faculty, or none at all.3 For these reasons it seems prejudicial to the debate to continue using the traditional translation of ‘free will’ for liberum arbitrium. Various alternatives have been suggested and used –‘free choice’, ‘free judgment’, ‘free decision’.4 I have chosen to use the single word ‘freedom’ to stand for the Latin phrase. Liberum arbitrium is often used by Aquinas as a synonym for libertas.5‘Freedom’ is an English term which can stand for both, and using it helps us to see that Aquinas is interested in a range of philosophical concerns which go beyond what might be suggested by a translation such as ‘free judgment’. Like the Latin phrase, it leaves open the question of how the human being is free and where that freedom lies; it emphasises neither an intellectual nor a volitional interpretation; and it indicates the subject of the argument and not its conclusion. I hope this will facilitate the development of ideas in this article.6

2. Freedom, choice, and preference

There are many passages in the Summa Theologiae concerned with freedom and choice – in relation to God, to angels, and to human beings. Freedom is not, for Aquinas, a third power which underlies or complements the work of intellect and will, it is the unified functioning of these two powers. Freedom is simply the working of intellect and will. We are free because we understand and desire. It will help to begin looking at one of the more systematic accounts. In Question 83 of Part I of the Summa Aquinas asks specifically about the nature of human freedom. He takes it for granted that human beings are free, appealing – as we might now say – to the witness of sociologists, lawyers, psychologists, and parents (‘otherwise counsels, exhortations, commands, prohibitions, rewards, and punishments would be in vain’).7 Then, by explaining how we are free, he also explains what freedom is. It is the fact that the human being, unlike other animals, ‘acts from free judgment and retains the power of being inclined to various things [potens in diversa ferri].’8 Freedom is thus our ability to seek different things because we can think about things in different ways. Actions are concerned with contingent, concrete matters, and ‘in such matters the judgment of reason may follow opposite courses [ad diversa se habet], and is not determinate to one. And forasmuch as human beings are rational is it necessary that they be free’.9

Aquinas takes up the definition of ‘what is free’ (liberum) from Aristotle's Metaphysics, cited in the third objection of the same article. ‘What is free is sui causa’ (‘cause of itself’ or ‘self-determining’).10 He clarifies this in his response. ‘Freedom is the cause of its own movement, because by their freedom human beings move themselves to act.’11 These themes are repeated in the first question of Part I-II, and one citation brings them together very succinctly: ‘Those things that have reason, move themselves to an end, because they have dominion over their actions through their freedom, which is the faculty of will and reason.’12

These, then, are some of the ideas associated with freedom: a judgment that is not determined, the ability to seek different things, the indetermination of reason, having control over one's actions, self-movement, self-determination. In the third article of I.83 Aquinas makes more explicit what is at the heart of each of these characterisations of freedom: choice. ‘The proper act of freedom is choice [electio]: for we say that we are free because we can take one thing while refusing another, and this is to choose.’13

Now two things concur in choice: one on the part of the cognitive power, the other on the part of the appetitive power. On the part of the cognitive power, deliberation [consilium] is required, by which one thing is judged [diiudicatur] to be preferred to another; and on the part of the appetitive power, it is required that the appetite should accept the judgment of deliberation.14

So there are two elements to any choice – a rational preference and a willing acceptance. Choice always involves a double movement: it is (A) judging what is to be preferred and (B) accepting the judgment. Or to put in another way, it is the unified movement of allowing the judgment of preference through its acceptance. It seems, to put it crudely, as if the cognitive power first does all the hard work of determining what is best to do, and then the will just rubber stamps this. We should not, however, be misled by the description of this sequence. It does not imply that the ‘choice’ of which path to follow is made solely by the judgment of deliberation, as if the intellect can always provide us with sufficient reason to prefer one course of action rather than another, and the will simply ratifies this irrefutable judgment. Aquinas is insistent that choice is a function of the cognitive and appetitive powers working together, and nothing is actually chosen unless the will accepts what is understood to be preferable. His reply to the second objection is highly nuanced.

Judgment [iudicium] is a sort of conclusion and termination of deliberation [consilium]. Now deliberation is terminated [determinatur], first, by the sentence [sententia] of reason; secondly, by the acceptance of the appetite: whence the Philosopher says in Ethics 3 that, ‘having formed a judgment by deliberation, we desire in accordance with that deliberation’. And in this sense choice itself is a kind of judgment [quoddam iudicium] from which freedom [liberum arbitrium, ‘free-decision’] takes its name.15

At first sight this text might seem to confirm the sequential description of understanding and willing introduced in the body of the article – the will (inevitably) follows the reason. Yet two enormously important qualifications are made here. First, deliberation, which might seem to be an independent rational process, is not in fact ‘terminated’ or ‘determined’ (determinatur) without the intervention of the will. In one sense it is still true to say that the reason determines the final deliberation, since (if the deliberation takes place) there is nothing apart from the sentence of reason for the will to accept. But on the other hand, without the concluding acceptance of the will, there is no deliberation, and the reason remains ineffective. In this sense, it is possible to say that the will determines the deliberation, since it determines whether any particular judgment of reason ultimately becomes effective. Deliberation is not complete (and therefore a preference is not made) until the will accepts the sentence of reason. So the function of the will is not simply to accept (or reject) the conclusions of deliberation, it actually plays a part in bringing deliberation to a conclusion about what is preferable. The integrated work of choice, which involves reason and will, is what brings the deliberation to a close. The second qualification made in this response is that choice (effected by the understanding and the will) is itself a kind of judgment, and it is not just the carrying out of a previous judgment made by the understanding in deliberation.

The background assumption to this article is that the sentence given by reason is inconclusive, which is why it can only be concluded and determined if it is finally accepted by the will. This is not true of all decisions. We often make a decision without choosing, if we rationally work out that one solution is clearly the best. Yet when we face a choice, we find that two or more options are acceptable according to the sentence of reason. They may be acceptable in different ways, but they are nevertheless both rationally acceptable. This is the very reason we have a choice. The ‘reason’ we have to choose is that there are no compelling reasons to act, or put another way, that there are too many conflicting reasons to act. We can think of trivial and serious examples: we have to choose between eating an apple or a pear, between watching the sport or the comedy on television, between giving oneself up to the police or escaping into lifelong exile, between forgiving someone or hating that person. In each case our reason can see the sense of each alternative action and may be unable to decide between them on rational grounds alone. We have to make a choice, which involves actually accepting one option when both are acceptable in theory; which involves actually preferring one option when both are preferable in theory.

These nuances from I.83 are apparent in the question about choice in Part I-II. Choice, Aquinas writes, involves both reason and will. ‘The will tends to its object, according to the order of reason, since the apprehensive power presents its object to the appetite.’16 This implies an orderly sequence, and might suggest that the reason compares the various options and determines which single option is preferable, as if there were a kind of rational necessity involved for anyone who is thinking properly. But the whole point is that we have to make choices when one option is not obviously better, when numerous options all make sense, and when each one could reasonably be chosen. John Finnis emphasises this and is highly critical of theories that might obscure the fact that for Aquinas choice is between rational, viable alternatives. ‘Any deliberation which ends in choice must have yielded, not one judgment affirming the choiceworthiness of an option awaiting adoption by the will, but (at least) two judgments.’17 Aquinas makes this clear in the following article when he writes that the will, unlike the sensitive appetite shared with animals, is ‘indeterminate in respect of particular goods’.18

Since choice is the taking of one thing in preference to another it must of necessity relate to several things that can be chosen [necesse est quod electio sit respectu plurium quae eligi possunt]. Consequently in those things which are altogether determinate to one there is no place for choice.19

Kevin Flannery draws attention to the fact that for Aquinas, and not for some of his neo-scholastic interpreters, there is still a choice to make even after the intellect has made all the judgments that it can:

If the process of practical reasoning truly leads to choice [electio], at the threshold of choice, there must yet exist options among which the agent chooses. The scholastic ordered pairings consilium-consensus/iudicium-electio suggests that the job of voluntas is to deliver propulsion (by consensus and electio) to what is decided only in intellect (consilium and iudicium). The genuinely Thomistic order, on the other hand –consilium, iudicium, consensus, electio– makes it apparent that the entire moral agent is present right at the very threshold of going into action.20

3. The inconclusiveness of reason

So freedom is associated with choice, and choice with indetermination. Aquinas thus has a very distinctive explanation of our freedom to choose: it derives from the fact that in practical matters reason itself is undetermined. Aquinas links freedom with the indetermination of reason in a number of key passages. In the first article of I.83, for example, he argues that our practical judgments are free and that we can incline ourselves to different goods because we are reasoning about particular, contingent things. This type of reasoning, like dialectical and rhetorical argument, does not lead to a single, scientifically demonstrable conclusion.

In such matters the judgment of reason may follow different courses, and is not determined towards one [iudicium rationis ad diversa se habet, et non est determinatum ad unum]. And insofar as the human being is rational is it necessary that the human being be free.21

Aquinas's view is striking: Practical reasoning about contingent things is necessarily inconclusive. So when, for example, we examine our lives and the situation before us, taking into consideration all the relevant facts, trying to work out what we shall do, we will always find that no single answer presents itself. The most meticulous analysis of all the available data, the most clear-sighted view of the issues involved, will be inconclusive and will leave us facing alternative courses of action. This will not be because we have missed something, it is part of the nature of paying full attention to the situation and thinking about it carefully. A single present allows for multiple possible futures; a single human situation allows for multiple possible actions.

The inconclusiveness of reason in practical matters concerns above all the determination of which good we shall seek and what our end should be. In I-II.13:6c Aquinas gives the reason unlimited flexibility in its ability to see particular things (but not the perfect good which is happiness) as desirable or not.

Now the reason can apprehend as good, not only this, ‘to will’ or ‘to act’, but also this, ‘not to will’ or ‘not to act’. Again, in all particular goods, the reason can consider the aspect of having some good, and the lacking some good, which has the aspect of bad [potest considerare rationem boni alicuius, et defectum alicuius boni, quod habet rationem mali]: and in this respect, it can apprehend any single one of such goods as to be chosen or to be avoided.22

This is a huge claim – that any concrete thing at all can always be seen as good in one way, or as not good. Choice is not just about those rare moments when we stand before two finely balanced and incompatible options. Aquinas says that we can always see more than one way of acting, because we can always see an aspect of good and an aspect of bad in any option, and therefore we can always discover reasons for doing it and reasons for not doing it. It is part of the nature of reason for Aquinas that it can observe present reality in different ways. The world has multiple possible meanings. The particular situations we encounter always and necessarily give rise to more than one conception of what is good, more than one practical option, more than one possible future. It can't be emphasised enough that it is reason which does this. Reason discovers that there is no necessity about any single interpretation of the good – this is not because of a failure of reason.

Aquinas addresses this question of necessity in the same article (13:6). The second objection is very forceful, and he accepts its argument: If there is a necessity about the prior judgment of reason then it seems there will be a necessity about the choice. In other words, if we have to think that something is good then we will inevitably choose it. In the body of his reply he meets this objection by restating his central contention. ‘The human being does not choose of necessity. And this is because that which is possible not to be, is not of necessity [quod possibile est non esse, non necesse est esse]’.23 We can choose without necessity, therefore, because of the possible not-being of the options. What makes them options is their conditionality, the fact that they could be and therefore that they are not (yet); the fact that they do not come about as a necessary consequence of the being of the world as it is. ‘Being’ (the reality of the present situation) gives rise to ‘possibility’ (the possible futures that are not yet determined) through the mediation of reason. When we face a choice we face options which precisely do not exist, they do not flow out of the present constitution of the universe with any necessity. These options have to be ‘made to be’ through the choice, and that is why it is not possible for there to be any necessity in the reasoning involved in a choice.

David Gallagher writes about how important it is for Aquinas that human beings are free to understand things in different ways and not just free to act:

Does the agent have control over how the options appear? This question cuts to the heart of the matter. If we say that choice and action depend upon how various goods appear to a person, and if a person does not control how these goods appear, then the person's action will not truly be free – able to be otherwise – nor will the person be morally responsible for it. If goods simply appear to an agent as they appear, then to characterize the will as rational appetite leads us into a form of psychological determinism, a determinism incompatible with freedom and responsibility.24

Gallagher explains that for Aquinas the agent exercises control over the very act of reason which governs his or her choice. ‘How objects appear, in terms of good or evil, is not simply a question of those objects taken independently of a particular agent, but rather depends in large measure on the agents themselves.’ An agent's ‘contribution’ to the appearances is always to some extent voluntary or willed.25

Aquinas's specific replies to the first two objections of I-II.13:6 contain two remarkable glosses on the nature of the type of reasoning that takes place in practical judgments. The first response runs:

The conclusion does not always of necessity follow from the principles, but only when the principles cannot be true if the conclusion is not true. In like manner, the end does not always necessitate in the human being the choosing of the means, because the means are not always such that the end cannot be gained without them; or, if they be such, they are not always considered in that light.26

One end can be achieved in different ways. One set of principles can lead the reason to a number of different conclusions. The reason by itself cannot ‘decide’ which of these reasons is to be followed, because they are all reasonable. The second response adds:

The reason's decision or judgment of what is to be done is about things that are contingent and possible to us. In such matters the conclusions do not follow of necessity from absolutely necessary principles [non ex necessitate sequuntur ex principiis necessariis absoluta necessitate], but from principles necessary only given a condition [sed necessariis solum ex conditione]; as, for instance, ‘If he is running, he is in motion.’27

So in these cases when there are many legitimate conclusions and multiple futures, the only way that a single conclusion is reached is when a condition is inserted which turns the principle into the kind of principle that requires a single answer. In other words, we have to create the conditions in which one conclusion will make sense.

Let's say that a woman has to choose whether to take option A or B in order to achieve the goal X. Let's accept that these are viable options, they arise from the reality of her situation and the possibilities available to her. X is the end, the guiding principle, which serves as the principle in a practical judgment. Perhaps she wants a salary and has to choose between being a teacher or a car mechanic; perhaps she wants a holiday and has to choose between a trip to Italy or Spain. She already knows that both options (A and B) lead to X – this is the very reason she has a dilemma. If she chooses A, all the specific benefits of A will accrue (together with the goal X); if she chooses B, all the specific benefits of B will accrue (together with the goal X). She cannot come to an unconditional conclusion on the basis of reason alone. She can only conclude that A is the correct conclusion if she first decides to build the conditions of A into the very principle X from which she is trying to derive A. A will be the conclusion that derives from her desire for X only if she sees the problem (and the solution) in terms of A, and decides to appreciate the specific benefits arising from A. There is a fundamental insufficiency about X.

This radical insufficiency of reason to come to a practical conclusion actually reflects a superabundance and not an insufficiency of viable options.28 It manifests itself when we find we have consented to more than one option:

It may happen that through deliberation several means have been found conducive to the end, and since each of these meets with approval, consent is given to each [in quodlibet eorum consentitur]; and from the many options that are approved, we give our preference to one by choosing it [sed ex multis quae placent, praeaccipimus unum eligendo].29

Aquinas could not be clearer here about the remarkable fact that we can approve of and consent to many options at the same time. ‘Since each of these meets with approval, consent is given to each [dum quodlibet placet, in quodlibet eorum consentitur].’ In these cases the reason alone is not able to find a preference. In fact, a preference is not something that is ‘found’, it is made or given to one option by choosing or in the very choosing (eligendo).

A reasoned analysis of the situation produces many possible courses of action. None of them arises from the facts before one with any inevitability, none of them makes a claim on us with any necessity. Reason alone is insufficient for determining our ends. We allow ourselves to be motivated by one set of reasoning by acting for a specific end, and this free choice to seek one end is what gives legitimacy to this specific set of reasons. In Aquinas's scheme it is reason that allows us to see the alternative possibilities for good within being, and the will that allows us actively to move ourselves towards one of these possibilities. The reason liberates us from necessity and the will recreates a kind of conditional necessity that is based on the freely chosen end.

In the view of Aquinas, we are free to act (in one way rather than another) because we are free to reason and to understand the good (in one way rather than another). An English phrase captures this nicely: the act of choosing is often called ‘making up one's mind’.30 The judgment of preference takes place in the very choosing, and one's will is one's capacity to shape oneself by responding to reasons.31 This means that in their choices human beings are freely deciding how they will understand the world, what they will prefer, and where their lives are going. We should note that Aquinas's account of choice does not just apply to those dramatic ‘Moments of Decision’ when we hesitate before an agonising dilemma that will determine the direction of our life and the quality of our character.32 Whenever an action is ‘up to us’,33 whenever we could have done otherwise, we then have to choose to do it. Both the seasoned Mafia hit-man and the loyal charity worker may go about their business without much reflection or hesitation, but they are still freely choosing to do their work and fully responsible for it. They could have done otherwise, if only by not acting.

Aquinas connects the fact that there are alternative contingent solutions to practical dilemmas with our ability to deal with universals. Human knowledge is not tied to particular, material things, and for this reason we can – as it were – direct and apply our ideas to various things by our free choice. If an architect had only a particular material form of a house in mind, one that was already individualised, then he would not be able to build any other type of house, since there would be only one concrete idea determining his thinking and motivating his actions. (For example, if an architect were using some off-the-shelf plans for a three-bedroomed, two-storey house made of wood and glass that had already been constructed a hundred times.) But architects usually start with universal forms (e.g., ‘a family home’, ‘an office block’) that can be realised in different concrete ways. Aquinas explains this in De Malo 6:

An intellectual form is a universal, under which many things can be comprehended [Forma intellecta est universalis, sub qua multa possunt comprehendi]. Hence, since acts are concerned with singulars, among which there is none that is equal to the potentiality of the universal, the inclination of the will remains indeterminately related to many things [remanet inclinatio voluntatis indeterminate se habens ad multa]; for example, if an architect conceives the form of a house in a universal, under which houses of different shapes are comprehended, his or her will can be inclined to build a house that is square or circular or of some other shape.34

The same building analogy is used in a question about the Divine will in De Veritate, but here Aquinas connects the universality of human reason not only with the architect's ability to embody universal ideas in different particular ways, but also with the architect's ability to decide whether to build the house or not.

Because the form of the house in the mind of the architect is the idea of the house taken absolutely [ratio domus absoluta], of itself not disposed any more to existence than to non-existence or to existence in one particular way rather than in another [magis ad esse quam ad non esse, nec ad sic quam ad aliter esse], as far as the accidental features of the house go, the architect's inclination in regard to making the house or not remains free.35

Once again, the future is not determined by the being of the present. The ratio absoluta of the house is not disposed more to existence or to non-existence, to one kind of embodiment or to another. So knowledge gives us an ‘indifference’ to being, an ability to decide whether something shall be or not be. The fact that we can abstract immaterial forms and think about things in general is what allows us to go beyond the totality of the present and envisage what does not have to be, which is another way of saying that we envisage what could be. Possibility and the conditional tense only emerge through this process of stepping back from concrete being (making a deeper sense of it, through knowledge) and stepping forward beyond concrete being (seeing the possibilities, through practical reasoning about human action). We don't just know that the future is open and undetermined – it is our knowledge which makes the future undetermined. Human knowledge introduces the potential being of multiple human actions through the actual non-being of the single understood form. Our ability to deal with universals which do not exist in concrete reality is what frees us from necessity and determination. Alan Donagan summarises Aquinas's view in this way:

[Freedom] is wholly a matter of the non-necessity of any judgment a man can arrive at by his natural powers as to the goodness of an end or the suitability of a means. Even when will seems to fly in the face of intellect, there is always a (foolish, perhaps vicious) judgment which directs it.36

Donagan points out that to his immediate successors Aquinas seemed to be affirming the priority of intellect over will, but as we shall now see there is a particular kind of priority that belongs to the will.

4. The influence of the will over reason

If reason itself cannot determine what is best to do, what does? If in questions of human action ‘the judgment of reason may follow opposite courses, and is not determinate to one’,37 what finally determines that a certain judgment be made? Aquinas believes it to be the will. One has to be extremely careful about the way this is phrased in order to avoid misinterpreting him. The will determines that a certain judgment be made while the reason determines the nature of the judgment actually made. The two faculties working together in this way constitute our freedom. We have seen that the acceptance by the will of a set of reasoning is what completes a choice. In practical matters, which are necessarily open-ended, we prefer something by willing one understanding of the good. Now in this section we can investigate this movement of the will that concludes the act of choice.

Aquinas touches on the activating power of the will in a number of articles. The will as an efficient cause [per modum agentis] moves the intellect and all the powers of the soul, ‘because wherever we have order among a number of active powers, that power which regards the universal end moves the powers which regard particular ends’.38 With respect to their exercise, the will moves the other powers of the soul to their acts, ‘for the end and perfection of every other power is included under the object of the will as some particular good’, and the will moves the other powers to their particular ends as it seeks the universal end.39 As to the exercise of its act, ‘no object moves the will necessarily, for no matter what the object be, it is in one's power not to think of it, and consequently not to will it actually’.40

Article I-II.10:2 contains perhaps the most unambiguous description of the decisive role of the will in determining human action. The question concerns whether the will is moved by its object of necessity. Aquinas writes:

If the will be offered an object which is good universally and from every point of view, the will tends to it of necessity, if it wills anything at all; since it cannot will the opposite. If, on the other hand, the will is offered an object that is not good from every point of view, it will not tend to it of necessity. And since lack of any good implies some non-goodness [quia defectus cuiuscumque boni habet rationem non boni], consequently, that good alone which is perfect and lacking in nothing, is such a good that the will cannot not-will it: and this is happiness. Whereas any other particular goods, in so far as they are lacking in some good, can be regarded as non-goods [alia autem quaelibet particularia bona, inquantum deficiunt ab aliquo bono, possunt accipi ut non bona]: and from this point of view, they can be set aside or approved by the will, which can tend to one and the same thing from different points of view [quae potest in idem ferri secundum diversas considerations].41

We can notice the following points: (A) If the will is actually in the process of willing, then it is not free not to seek the perfect good in happiness. So there is a general necessity about willing our final end. (B) Apart from the perfect good, absolutely any other good at all can be viewed as good or as not good. In other words, even though Aquinas insists that the object specifies the act and the reason determines what is good, nevertheless the object and the reason alone can never ‘specify’ which specification of the good will motivate the act. Reason supplies too much information – it can never present the will with a single, indisputable possible good (apart from the perfect end). (C) Particular goods can be either set aside or approved by the will (possunt repudiari vel approbari a voluntate). This is crucial. Even though Aquinas sometimes simplifies his account and suggests, as we have seen, that the cognitive power alone (through deliberation) judges what is preferable,42 nevertheless deliberation itself is not terminated without the acceptance of the will,43 and a preference cannot be given without the affirmation of the will that closes choice.44 (D) When the will sets aside or approves a particular object, when it accepts one good rather than another, this is because it sets aside or approves of a point of view which is reasonable, which is one legitimate way of understanding this object. So the will is not going against reason; rather the will is selecting one reason from amongst many. The act is still specified solely by the goodness of the object as presented by reason. It is the object which moves and determines the act as its specifying principle, the object which makes it this act and not another.45

In an article about choice from De Veritate Aquinas is slightly fuller in his description of the distinctive role of the will. He is discussing the nature of choice:

Choice is the final acceptance [ultima acceptio] of something to be carried out. This is not the business of reason but of will; for, however much reason puts one ahead of the other, there is not yet the acceptance of one in preference to the other as something to be done [nondum est unum alteri praeacceptatum ad operandum] until the will inclines to the one more than to the other. The will does not of necessity follow reason [non enim voluntas de necessitate sequitur rationem]. Choice is nevertheless not an act of the will taken absolutely but in its relation to reason, because there appears in choice what is proper to reason: the putting of one next to the other or the putting of one before the other [conferre unum alteri, vel praeferre]. This is found in the act of the will from the influence of reason: reason proposes something to the will, not as useful simply, but as the more useful to the end.46

Once again it must be emphasised that the reason is not proposing a single reasonable plan of action which is automatically approved by the will. Sometimes alternative plans of action cannot be ordered by the reason, they are all equally reasonable, and the reason ‘puts them next to one another’ (conferre). Sometimes the reason puts one plan before the others (praeferre) – but without losing sight of the viability and reasonableness of the alternatives. In neither case is the will obliged to prefer one alternative as ground for action (ad operandum) rather than another. The inclination of the will to one rational plan, which is the ultimate cause of action, is not necessarily determined by the order given by reason. There couldn't be a clearer statement of the determining influence of the will. Aquinas's own words, however, could mislead us here. When he writes that the will does not of necessity follow reason he means here that it does not follow the ordering (praeferre) which reason gives amongst rival plans. It still, however, follows the reasonableness of the chosen option. As he goes on to say, choice is always an act of the will in relation to reason, and the option preferred by the will is always therefore a reasonable one that has been proposed (even if it is a lower ranking proposal) by reason.

Stephen Brock draws attention to the role of the will in making the preference.47 When we reason about possible actions, ‘it can happen that both “measure up”and that neither is a clear winner’. If taking one excludes the other, then the decision is simply ‘up to you’.

You refuse one when all conditions needed for your accepting it are present, and you accept the other when all conditions needed for refusing it are present. […] The will moves toward one thing despite a sufficient attraction toward something excluding it; this is a choice, a taking one thing in the face of another, a preference. This is not at all to say that the choice is not informed by a judgment. The chooser must have formed a judgment declaring the preferability of what he chose. Only, he also formed a judgment declaring the preferability of the alternative he rejected. Nor are these two separate judgments; it is one judgment, declaring one alternative preferable in some respect, and the other preferable in another respect.48

So there are multiple practical truths, multiple possible acts, which all make sense in different ways. Let's call them different lines of reasoning. Reason cannot decide between them, since reason is the very faculty that has brought them to light. It is up to the will to prefer one way of reasoning and acting. This is free choice. It is simply the way that we activate a reason. Note that it does not involve an additional, alternative, non-rational apprehension of good. Stephen Brock puts it this way:

[The will is not] an additional source of objects or specificatory principles, outside or apart from those given by the intellect. Rather, the will plays a role in the determination of its object precisely by playing a role in the process by which the intellect comes to provide it with an object.49

The will is not determining what is good (the reason does this), it is determining that one way of looking at one good should be activated, that one project be followed. The only reason for doing X is X itself – as judged by the reason. The movement of the will is necessarily in accord with the good as it is presented by reason, but the movement itself is not caused by the understanding of the good – it is caused by, indeed it is, the will's very attraction to this good. A choice is rational, indeed there is no such thing as an irrational choice (since it must be between reasonable options) – yet a choice is not rationally made.

This whole approach to understanding the relationship between intellect and will in Aquinas has been questioned by a number of recent commentators, who follow a more ‘intellectualist’ line.50 Broadly speaking, they deny this so-called ‘voluntarist’ interpretation, in which the will has some kind of final control over its own activities, and they insist instead that its activities are controlled by the intellect. So in the particular situation of choosing between alternative rational goods it is not the will that inclines us to one good (or to one consideration of the good) rather than to another; or if it is the will, then this will is following the conclusions of the intellect about which good is to be rationally preferred.

This is not the place for a line-by-line rebuttal of the ‘intellectualist’ stance. I hope that my own reading of the texts has shown that one can deny Aquinas is an intellectualist without turning him into a voluntarist, insofar as the good eventually chosen in always completely specified solely by the intellect. In my understanding of Aquinas, the will is not against reason, it is what establishes it as something with practical relevance. Why did someone act according to one reason rather than another? Why did we go to the cinema instead of the bowling alley? Eat Italian instead of Mexican? Talk about football instead of politics? On the one hand, the action brings about its own explanation – we act for the objective good sought. The primary reason for eating Italian food is because Italian food is good. But why do we not follow other reasons? Because of the reasonableness of the reasons that we do choose. There are often no further reasons to put into the equation. This is not voluntarism, for the simple fact that the movement of the will towards this good is explained by the objective rationality of the good in question. The fact that other goods could also have been reasonably chosen does not alter this. On the other hand, there is no explanation beyond the freedom of the one who acted.51 The goodness of Italian food is not enough to explain the choice since Mexican food is equally good. We freely determine ourselves to act in this way, to follow these reasons. I do this because I choose to: that is the reason. There is something irreducible about the movement of the will that results in a choice being made. It is a kind of unanalysable fact. A choice creates something new. This is still not voluntarism, since the movement of the will is never made against reason or in isolation from reason – it is the very thing that allows me to follow my reason in this way.

By choosing one way of reasoning we are giving priority to one notion of good and orienting our life to one goal. John Finnis gives an example of a scholar dedicated to the pursuit of the truth who abandons this for a new cause such as fighting for his community or caring for his sick wife. His new commitment has not somehow become more reasonable; rather, the change in his chosen life-plan has made the reasonableness of this new commitment more persuasive.

That chosen plan made truth more important and fundamental for him. His new choice changes the status of that value for him; the change is in him. Each of us has a subjective order of priority amongst the basic values.52

So Aquinas is neither an intellectualist nor a voluntarist. He believes that we creatively determine which understanding of the good will motivate our personal actions even though each understanding is determined solely by the nature of the good as understood by reason. In other words, the objectively understood good can be deeply personal. This is the kind of moral synthesis so many contemporary thinkers struggle towards. Charles Taylor, for example, is acutely aware that some ethical theories can depersonalise human action. Choices are based on the objective values that our disengaged reason discovers. Yet he knows that other theories which appeal to the language of self-realisation and subjective fulfilment run the risk of losing sight of the objective good. Caught between rationalism and romanticism, Taylor pursues a ‘search for moral sources outside the subject through languages which resonate within him or her, the grasping of an order which is inseparably indexed to a personal vision’– he wants value to be both objective and subjective.53 Aquinas described just this ‘resonance’ in the language of intellect and will. The objective good must be subjectively accepted; the reasonableness of a right action must be personally willed by the agent. Aquinas's action theory thus allows for a concept of human autonomy which does not separate personal responsibility from a rational understanding of the objective good.54

5. Conclusion: The self-movement of the will and the self-constitution of the human person

When the will prefers one rational plan of action by inclining to it, there is no prior reason for this inclination (beyond the reasonableness of the good sought), it is the very exercise of our freedom. ‘The proper act of freedom is choice: for we say that we are free because we can take one thing while refusing another; and this is to choose.’55 Aquinas's explanation for this is simple, and it is more a description than an explanation: the will moves itself. The will can ‘pass or not pass into the act of willing with regard to anything at all’ because ‘animate things are moved by themselves’[moventur a seipsis].56 Self-movement gives us control over our actions and independence from the totality of causes which press upon us.

Those things that have reason, move themselves [seipsa movent] to an end, because they have dominion over their actions through freedom, which is the faculty of will and reason.57

Self-movement belongs properly to the will and not to the intellect, since the intellect is moved by the will to act, ‘but the will is not moved by another power but by itself’.58 The idea of self-movement implies that the fact of movement has no cause outside the occurrence of the movement itself. Aquinas states that ‘freedom is the cause of its own movement [causa sui motus], because by one's freedom one moves oneself [seipsum movet] to act’.59

De Malo 6 is particularly helpful on this question of the will's self-movement.60 Aquinas is writing about the exercise of the act (i.e., whether an end that is already understood by the reason in a particular way will actually be pursued), rather than the specification of the act (i.e., whether an end will be understood by the reason in one way rather than another).61 He explains that ‘the will is moved by itself [voluntas movetur a se ipsa]: for just as it moves the other powers, so also does it move itself [se ipsam movet]’.62 He recognises that this seems to imply a contradiction, since moving something normally involves one thing that is not in motion being moved by another thing that is in motion. How can the will be both not in motion and in motion at the same time? He writes that in this case the will is not ‘both in potency and in act with regard to the same thing’.63 For just as our knowledge of one thing leads us on an investigation that results in some new knowledge, so the fact that we already will one thing (such as health) leads us to will another thing (such as the taking of some medicine).

Aquinas then makes two striking clarifications. First, we only will a particular means (such as medicine) if we are willing to take counsel (consiliari) about how to achieve an already established end (such as health). And given that the will moves itself by counsel, and ‘counsel is a kind of investigation that is not demonstrative but involving opposites, the will does not move itself of necessity.’64 So the lack of necessity, the freedom, flows from the fact that a will that is already willing a certain end can continue willing that end in different rationally valid ways. The willing of a concrete good (such as medicine) is thus never a new and self-generating act, it is always part of an already established movement toward some greater goal. This is what preserves both its freedom and its rationality.

The second clarification is about the transcendent cause of the will's overarching movement.65 Aquinas writes that even the act of taking counsel must be willed, and that this act of will requires its own act of taking counsel, which seems to lead to an infinite regression. Aquinas concludes that the will must be moved ‘by something external, by the impulse of which the will begins to will [ab aliquo exteriori, cuius instinctu voluntas velle incipiat]’.66 Given that the rational soul is immaterial, this initiating force cannot be material – it must be something above the will and the intellect, namely God.67 But in this case God ‘moves the will according to its condition: not from necessity but as indeterminately relating to many things [voluntatem movet secundum eius conditionem, non ex necessitate set ut indeterminate se habentem ad multa]’.68 So the will is not the cause of its own initial or originating movement. Nevertheless, the will is moved according to its ‘condition’ or ‘nature’[conditio], which is to be open to many things in a way that is indeterminate. It is, to use a slightly strained phrase, necessarily indeterminate. In other words, the transcendent foundation of the will (in God) does not take away from its freedom to move itself to different possible goods. God makes the will to be what it is, which is an inclination to happiness that can be embodied and fulfilled in many different ways – and the decision about which way depends on the human person and not on God.

Why, finally, does one choose to do this rather than that? As we have seen, Aquinas believes that the choice doesn't depend on something else; rather it is self-constituting. David Gallagher formulates this in a startling way:

The judgment of choice which determines the will's motion arises in the choice, a choice which occurs only when it is willed. Hence the will influences, in the act of choice, the very judgment it follows in that act.69

There is no actual circularity here, since choice is a single human act of an individual person. In that one act, by means of two powers, we determine ourselves to a particular action and establish that one judgment of the good (out of many possible judgments) will govern that action.70 There is a kind of ‘immanent’ operation here, which must, as Stephen Brock explains, ‘be an activity which is immediately and simultaneously able to effect either one thing or its contrary’.71

It is no exaggeration to say that self-movement is a kind of self-creation, since the decision to move towards a certain end defines the person one becomes. When we face a choice about how to think and what to seek, whether small or large, we face a choice about ourselves. Before the moment of choice, our reason discovers that there are different ways of interpreting the matter in hand, different ways of understanding our good, all of which arise from the present reality of the world and of our self. So our identity, which is formed in part by our understanding and by the goods we seek, is in question. Our reason has discovered that there is no single objective way of understanding things, no single set of necessary practical goals, no fully determined ‘self’ to guide us. The self is open-ended. Then, at the moment of choice, our will approves of one specific way of understanding things by actually seeking one specific good. We go beyond the indeterminate world of possibility and move towards a single determined future. So our personal identity, which is constituted in relation to the goods that we actually seek, becomes established. We re-create ourselves by seeking a particular form of perfection in a particular good. This is the sense in which Aquinas believes that we constitute ourselves through our free choices.


  • 1

    I will concentrate on some texts from the Summa Theologiae, De Veritate, and De Malo. I will not pay much attention to the differences between Aquinas's earlier and later writings on intellect and will. There is an ongoing debate about the development of Aquinas's thinking. I accept Daniel Westberg's conclusion that the differences are in emphasis rather than in matters of substance. Westberg argues that Aquinas did not, as Odon Lottin has proposed, move from a kind of intellectual determinism (in De Veritate) to a more voluntarist conception of human action (in De Malo 6). There is instead a consistent picture of the interdependence of intellect and will, a picture in which every action takes place for the sake of a good that is both understood and desired. Westberg writes: ‘Free choice is a matter of choosing, on the part of both reason and will, the bonum intellectum. This never changes in Thomas.’ Daniel Westberg, “Did Aquinas Change His Mind About the Will?”, Thomist 58 (1994): 56. There is, furthermore, an ongoing debate about the order in which the main texts concerning freedom were written, which complicates the discussion about Aquinas's intellectual development. Kevin Flannery argues that De Malo 6 is a much earlier work than usually thought, from no later than 1259, and that parts of De Veritate 24:1 are in fact based on De Malo 6. See Kevin L. Flannery, Acts Amid Precepts (Edinburgh: T&T Clark, 2001), 247–49.

  • 2

    The term goes back to classical literature and legal formulations where it indicates the ‘power to decide’ or ‘freedom of action’. See Daniel Westberg, Right Practical Reason: Aristotle, Action, and Prudence in Aquinas (Oxford: Clarendon Press, 1994), 81–82, and Charles H. Kahn, “Discovering the Will: From Aristotle to Augustine,” in The Question of ‘Eclecticism’: Studies in Later Greek Philosophy, ed. John M. Dillon and A.A. Long (Berkeley and Los Angeles: University of California Press, 1988), 250.

  • 3

    See J. B. Korolec, “Free Will and Free Choice,” in The Cambridge History of Later Medieval Philosophy, ed. Norman Kretzmann, Anthony Kenny, and Jan Pinborg (Cambridge: Cambridge University Press, 1982), 630–34.

  • 4

    See, for example, Westberg, Korolec, and also Timothy Suttor in Thomas Aquinas, Summa Theologiae, ed. Thomas Gilby, 60 vols. (London: Blackfriars: Eyre and Spottiswoode, 1963ff), vol. 11, 237, footnote a.

  • 5

    See, for example, Thomas Aquinas, Summa Theologiae I.83:1ad3; hereafter referred to without title. The Latin text is from the Leonine edition of Aquinas's complete works, that is, Sancti Thomae Aquinatis Doctoris Angelici Opera Omnia (Rome, 1882-), Volumes 4–11. The English translation in this article is based on that found in St. Thomas Aquinas, Summa Theologica, translated by Fathers of the English Dominican Province, (Westminster, Maryland: Christian Classics, 1948), 5 volumes.

  • 6

    This does not rule out the fact that there are other senses of freedom for Aquinas which lie outside the range of meanings included in liberum arbitrium and libertas; such as the free will (libera voluntas) that inclines us to our final end, even though it excludes any choice and involves a kind of natural necessity. Cf. Thomas Aquinas, Quaestiones disputatae de veritate, 24:1ad20, hereafter referred to as DV. The Latin text is from the Leonine edition of Aquinas's complete works, that is, Sancti Thomae de Aquino Opera Omnia (Rome, 1882-), Volume 22, Parts 1–3. The English translation in this article is based on that found in Thomas Aquinas, The Disputed Questions on Truth, translated by Robert W. Mulligan, James V. McGlynn and Robert W. Schmidt, (Chicago: Henry Regnery Company, 1952), 3 volumes.

  • 7


  • 8


  • 9


  • 10

    I.83:1obj3. Citing Aristotle's Metaphysics 1:2, 982b25.

  • 11


  • 12

    I-II.1:2c. He is citing Peter Lombard, II Sent., 24, 3. Cf. I.83:2obj2.

  • 13


  • 14


  • 15

    I.83:3ad2. Citing Aristotle's Nicomachean Ethics 3:3, 1113a12.

  • 16


  • 17

    John Finnis, “Object and Intention in Moral Judgments According to Aquinas,”The Thomist 55 (1991): 5–6.

  • 18


  • 19


  • 20

    Flannery, Acts Amid Precepts, 163.

  • 21


  • 22


  • 23


  • 24

    David M. Gallagher, “Free Choice and Free Judgment in Thomas Aquinas,”Archiv für Geschichte der Philosophie 76 (1994): 248.

  • 25

    Ibid.: 249.

  • 26


  • 27


  • 28

    This is one reason why Yves Simon insists that the key to Thomistic freedom is superdetermination and not indetermination. Yves R. Simon, Freedom of Choice (New York: Fordham University Press, 1969), 152–53.

  • 29


  • 30

    See Joseph M. Boyle, Germain Grisez, and Olaf Tollefsen, Free Choice: A Self-Referential Argument (Notre Dame and London: University of Notre Dame Press, 1976), 13.

  • 31

    See John Finnis, Aquinas: Moral, Political, and Legal Theory (Oxford: Oxford University Press, 1998), 66–70.

  • 32

    See Flannery, Acts Amid Precepts, 162–66. The examples that follow are based on Flannery's.

  • 33

    See Aristotle, Nicomachean Ethics, 3:5, 1113b6.

  • 34

    De Malo 6c [287–296], Thomas Aquinas, “Quaestiones Disputatae De Malo,” in Sancti Thomae De Aquino Opera Omnia (Rome: 1882-); hereafter referred to as DM. The English translation is from St. Thomas Aquinas, On Evil, translated by Jean Oesterle (Notre Dame, Indiana: University of Notre Dame Press, 1995). The body of DM 6 (a single article) is very long, and for this reason, instead of just referring in the customary manner to DM 6c [corpus], I also provide line numbers for each quotation in square brackets. These refer to the line numbering in the above Latin edition.

  • 35

    DV 23:1c.

  • 36

    Alan Donagan, “Thomas Aquinas on Human Action,” in The Cambridge History of Later Medieval Philosophy, ed. Norman Kretzmann, Anthony Kenny, and Jan Pinborg (Cambridge: Cambridge University Press, 1982), 652–53.

  • 37


  • 38


  • 39


  • 40


  • 41


  • 42


  • 43


  • 44

    I-II.13:2c and I-II.15:3ad3.

  • 45

    The object of the intellect is ‘universal being and truth’ (ens et verum universale), as Aquinas writes in I-II.9:1c.

  • 46

    DV 22:15c.

  • 47

    He is commenting on I.83:3.

  • 48

    Stephen Brock, Action and Conduct: Thomas Aquinas and the Theory of Action (Edinburgh: T&T Clark, 1998), 170, footnote 75; italics in original.

  • 49

    Ibid., 170, see 61–72; italics in original.

  • 50

    See, for example, Jeffrey Hause, “Thomas Aquinas and the Voluntarists,” Medieval Philosophy and Theology 6 (1997); P. S. Eardley, “Thomas Aquinas and Giles of Rome on the Will,” The Review of Metaphysics 56 (2003); and Robert Pasnau, Thomas Aquinas on Human Nature (Cambridge: Cambridge University Press, 2002), esp. section 7.4, 221–233.

  • 51

    See Gallagher, “Free Choice and Free Judgment in Thomas Aquinas,” for a particularly fine account of all these issues.

  • 52

    John Finnis, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980), 93, see ch. IV, 81–97.

  • 53

    Charles Taylor, Sources of the Self: The Making of the Modern Identity (Cambridge: Cambridge University Press, 1989), 510.

  • 54

    This idea of autonomy in Aquinas is brought out in Martin Rhonheimer, Natural Law and Practical Reason: A Thomist View of Moral Autonomy (Fordham University Press, 2000), see esp. viii and 143.

  • 55


  • 56

    DV 22:6c.

  • 57

    I-II.1:2c, citing Peter Lombard, II Sent., 24, 3.

  • 58

    DM 6ad10.

  • 59

    I.83:1ad3. But this doesn't exclude God being the first cause of our freedom, as we shall see.

  • 60

    See DM 6c [360–415].

  • 61

    But it is important to remember that the will's control over the exercise of an act also has some bearing on the act's specification, since any specification depends on a particular act of the intellect which itself needs activating.

  • 62

    DM 6c [361–363].

  • 63

    DM 6c [364–365].

  • 64

    DM 6c [378–381].

  • 65

    Cf. I-II.10:4 and I-II.109:2ad1.

  • 66

    DM 6c [390–391].

  • 67

    Aquinas refers to the conclusions in the chapter De bona fortuna of Aristotle's Eudemian Ethics, 8:2, 1248a16–29. Aquinas's whole theory rests on this Aristotelian assumption that there is an ‘external’ or ‘transcendent’ source of the will's dynamism.

  • 68

    DM 6c [412–415].

  • 69

    Gallagher, “Free Choice and Free Judgment in Thomas Aquinas,” 256.

  • 70

    Ibid.: 276.

  • 71

    Brock, Action and Conduct, 40, footnote 79.