You are here

21: The Concept of Will

PART III: Emotional Conflict and Structure of the Mind
21: The Concept of Will
I have been discussing the division of the will into two wills. But the will is itself a division of the soul and we need to consider the concept more closely. It evolved gradually and there have been many suggestions about who first formulated it. Suggestions have included Plato1 Aristotle2 the Stoics Chrysippus and Posidonius followed by the Platonist Galen3 the Stoic Seneca4 the Stoic Epictetus5 Augustine6 and Maximus the Confessor.7

To start very briefly with the terminology of the will it developed somewhat independently of the concept and both developed only gradually until their full flowering in the Latin of Augustine whose On Free Choice of the Will was written in AD 388–95. As regards terminology Aristotle sometimes reflects the practice of Plato's Academy which used boulēsis as a term for rational desire for the good as opposed to thumos for the desire for honour and epithumia for the desire for pleasure.8 Boulēsis both in this context and in others is often translated ‘will’.

It has been shown that in the Christian era forms of another word for willing thelein became more prominent.9 Thelein and thelēma are often used in the New Testament and those along with thelēsis in the Septuagint version of the Old Testament. When Christ asks for the cup to pass from him but nevertheless for his Father's will not his to be done the verb used is thelein more often than boulesthai and the noun is thelēma.10 Origen discussing human self-determination (autexousion) asks if it is threatened by Paul's remarks that reward does not depend on the man who wills (thelei) but on God's mercy or that it is God who wills (thelei).11 Among the pagans Epictetus uses the verb thelein often enough but not the corresponding nouns which are not common pagan usage although we shall notice thelēma for the will of the One in Plotinus12 Porphyry speaks of the soul's ethelousion13 and thelēsis appears once as a species of boulēsis for the Stoics.14 The Christian who made thelēsis the standard word for will it has been said15 was Maximus the Confessor in the seventh century.
But the most important terminological developments were in Latin. The phrase ‘free will’ libera voluntas appears in Latin in the first century BC in the Epicurean Lucretius followed by Cicero.16 But in Lucretius although there is an important discussion of freedom due to the unpredictable swerve of atoms the fact that this is connected with will does not prove to be very significant. The innovation I believe has more influence on terminology than on concepts.
As others have shown the Christian Tertullian writing in Latin shortly after AD 200 uses the phrase ‘free power of choice’ (libera arbitrii potestas) and ‘freedom of choice’ (arbitrii libertas).17 These are at least sometimes translations of the completely different Greek term to autexousion self-determination which makes no reference to choice or will.18 It has been suggested that the phrase ‘free choice of the will’ (liberum arbitrium voluntatis) originates with Augustine who uses it extensively in his On Free Choice of the Will.19 >From this account of Latin terminology it looks as if Boethius is reading Augustine's Latin expression back into an earlier Greek debate when he talks of free choice of the will.20
The concept: a history of clustering
But that is enough on terminology. I want to focus rather on the development of the concept and I want to reframe the central question. Instead of asking who invented the concept of will I think it is more profitable to ask something different since there is no one concept and much less do we have an agreed concept nowadays. Rather will is a desire with a special relation to reason and a number of functions associated with it. Some of these functions come in clusters. It is more illuminating to ask when these functions came together and who made the decisive difference.
The functions include two important clusters freedom and responsibility on the one hand and will-power on the other. My claim will be that both these clusters can be found early in Greek philosophy and even in the same philosophical treatise but totally dissociated from each other and often connected with reason rather than with rational desire. It is a long time before all the elements get associated together.
When they do get associated yet other ideas previously instantiated in isolation join the group: the idea of perverted will and of will as ubiquitously present in all decisions. Once this history of clustering is clear it will be the history that matters. As to when the concept of will was invented we can say what we like but we shall see the reasons for saying one thing rather than another.
At the end I shall explain why I do not think the invention has to wait for Maximus the Confessor.
Reason versus rational desire for the good: Aristotle's restructuring of Plato
Aristotle gave a strong impetus to the idea of will as a desire so distinct from reason but none the less belonging with reason as rational. In two passages Aristotle treats boulēsis as belonging to the rational part of the Platonist soul.21 At the same time in one of the two passages he urges against Plato that if one is going to distinguish parts of the soul one should bring boulēsis together with other types of desire thumos and epithumia to form a desiderative part of the soul (orektikon) quite distinct from reason.22 So far this seems to encourage the view that boulēsis is not reason but a rational desire. And I think this is close to Aristotle's view but I must enter a caveat. For Aristotle qualified the Platonist view that boulēsis belongs to the rational part of the soul as indeed Plato himself had done before him.23 At the opposite extreme Aristotle once says that boulēsis exists in children before reason or intellect (logismos nous) and is irrational.24 But elsewhere he calls it rational.25 And his more considered view is that the part of the soul that desires even if called irrational does have a share in (koinōnein) reason and can even in a secondary sense be said to have reason (logon ekhein) because it listens to reason even if it does not reason things out for itself.26 The general effect is to make boulēsis distinct from reason though still related to it.27
As regards boulēsis being directed to the good Aristotle draws on an idea found in three passages of Plato28 now thoroughly familiar from the preceding chapter where Plato was seen to use the verb boulesthai to say that what we really want is good. The contrast in each of the three passages is with mere appetite (epithumia) which is for pleasure not for good. Plato added we saw that no one is satisfied (arkein) with apparent good.29 But in Aristotle's version boulēsis is directed to what is or appears good.30 Later we shall see the Stoics had more to say about whether the good willed in boulēsis is real or apparent. The term ‘good’ (agathon) in this context is used by Plato and Aristotle in a narrow sense in contrast with honour and pleasure which are the goals of the lower types of desire thumos and epithumia.31 In some sense however Plato and Aristotle are ready to say that all desire not just boulēsis sees its objective as good in some way or other.32
Plato: freedom and responsibility separated from will power
Plato might seem an unpromising source for a concept of the will because we have already seen Aristotle criticizing him for not distinguishing sharply enough between reason and rational desire. None the less the good point has been made that something very like the function of will-power is assigned by Plato to another part of his soul high spirit or the spirited part (thumos thumoeides).33 Thumos is like will in being distinct from reason but a desire which according to Plato is always allied with reason and never opposes it in a struggle against appetites (epithumiai) although elsewhere it is sometimes shown opposing reason.34 Plato thus foreshadows the debates that arose in the 1270s AD as to whether will is free to oppose reason.35 But according to Plato's first-mentioned view thumos shepherds the baser appetites as if it were reason's sheepdog. So far it seems to play the role of will-power. What is missing is any particular connection with moral responsibility or with freedom so only some of the criteria for a concept of will are satisfied.
In another part of the Republic however choice (haireisthai) is connected both with freedom and with moral responsibility.36 Souls are represented as choosing their next lives before reincarnation. Because of the choice the responsibility will be theirs not God's:
Responsibility (aitia) is the chooser's; God is not responsible (anaitios).
Moreover they may choose virtuously and virtue is free: it has no master (adespoton). This is the earliest use I have encountered of the metaphor of freedom. It is earlier than the use of the same word by Epicurus to which Charles Kahn has drawn attention.37 What is still missing is any cross-reference to the treatment of thumos as will-power elsewhere in the Republic.
Platonists: the separation continues
The separation of the two subjects continues in later Platonists although they develop one of the subjects. They pick up Plato's term for freedom (adespoton)38 and integrate it much more fully with the ideas of responsibility and will and Christians follow. The Middle Platonist Didaskalikos says that if virtue has no master (adespoton) it must be voluntary (hekousion) and that since the soul chooses (helesthai) its next life and has no master it is up to it (ep’ autēi) whether it acts or not. Plotinus connects the term adespoton not only with choice (helesthai) and with what is up to us (eph’ hēmin) and voluntary (hekousion) but also with boulēsis39 one of the words conventionally translated as ‘will’. Virtue is up to us and without a master if we will and choose. Plotinus has an extended discussion in the treatise which Porphyry calls On the Voluntary and the Will [thelēma] of the One.40 We find extra terms not only for the will (thelēma) and willing (thelein) but also for freedom (to eleutheron) control (kurios) and purpose (proairesis). The Christian Gregory of Nyssa repeats that virtue and the soul have no master (adespoton) adds that virtue is voluntary (hekousion) and connects this with the self-determination (autexousion) of the human will (proairesis) or soul and with the soul being steered by its own willing (thelēmata).41
None the less Plato's idea of thumos as will-power which we found in an earlier part of the Republic is not integrated with these other ideas either by him or by later Platonists. When the power of thumos is treated by a later Platonist Galen it is again treated separately from these other ideas. Galen discusses contests of strength between reason (logismos) and high spirit (thumos). The talk is of strength (iskhuron rhōmē) violence (sphodroteron) domination (arkhein kratein epikratein) carrying off (sunapopherein) dragging (sunepispān) defeat (nikāsthai) and weakness (arrhōstia).42 In another passage it is all three elements of the Platonic soul that are involved not only reason and high spirit but also appetite (epithumia). Each of these can stir up or stop impulses (hormai). But already there is an important difference from the passage in Plato's Republic. Thumos in Galen is not always the ally of reason but frequently opposes it. So if we want to find rational desire in Galen's account we must look rather to his boulēsis boulēthēnai. But boulēsis is not treated in this context as having any special dominance over the other desires thumos or epithumia. Any of the three elements will have more or less power at different times.43 When Galen wants to give the highest element in the soul a special status comparing it with a charioteer or a rider who ought to take control he reverts to Plato's usage calling it reason (logismos) rather than boulēsis.44
Galen's discussion of will-power if it is one is not connected by him with any discussion of freedom or responsibility He does also discuss as others have well shown45 the power exerted from the brain through the nerves. But this is a discussion of physiological power and so takes us away from will-power.
Aristotle's proairesis distinct from responsibility and will-power
I have ascribed Aristotle a role in developing the concept of will in so far as he contrasted two concepts borrowed from Plato those of reason and of boulēsis or rational desire for the good. But it was boulēsis which I stressed. It is in Aristotle's other concept of proairesis that some interpreters have detected a concept of will.
Proairesis is generated from boulēsis which as Plato had already hinted46 is the desire for ends. Proairesis is the desire for the means which will lead towards those ends.47 And proairesis is even more closely connected with reason because it is a desire based on reasoning out what means would secure those ends.48 An example of proairesis I take it would be the kind of dietary policy which Aristotle cites like eating dry food.49 This is something one might have reasoned would lead to the goal of health.
I do not think the concept of proairesis as it features in Aristotle is yet very close to a concept of will. Aristotle does not treat proairesis as a kind of will-power. When he discusses people who fail to abide by their proairesis50 he does not present this as due to their proairesis being weak. In his main account of this failure of control (akrasia) he insists so I claimed in the last chapter that appetite makes us overlook the facts e.g. the fact that the food we are taking is not of the right sort.51 Furthermore he follows Plato and says that it is reason (logos) against which the appetites fight.52 He does not say it is will. Conventionally this discussion is spoken of as Aristotle's explanation of weakness of will. But this is a misnomer for the reason given that his diagnosis is not in terms of proairesis being weak or strong. If he concedes anything to the idea of will-power then I would agree with the point that has been put to me53 that it is when he discusses the opposite phenomenon that of maintaining control (enkrateia) and sticking by our proairesis in the face of rival desires. I agree that this is described in terms of winning or losing (nikān hēttasthai) against strong (iskhuros) desires.54 But the discussion is extremely brief because all the emphasis is given to failure of control. Moreover it is never said what enables proairesis to win when it does. But if Aristotle had addressed the question at least part of his answer would surely have been in terms of intellect rather than will. It must make a difference how carefully you have thought out your policy (proairesis) at the stage when you were deliberating about the best means to your goal. Aristotle's very silence is significant: a proponent of will-power would be likely to tell us how victory depends rather on the strength of the will.
Aristotle does not associate proairesis very closely with freedom and he even dissociates it from moral responsibility. He is explicit that voluntariness which he has just linked with moral responsibility55 extends much more widely to the doings of animals and children who are not capable of anything as rational as proairesis:
Although proairesis (deliberate choice) appears to be voluntary then it is not the same thing. The voluntary extends further. For both children and animals share in the voluntary but not in deliberate choice.56
I have elsewhere resisted an alternative interpretation according to which Aristotle distinguishes ‘up to us’ (ephhēmin) from voluntary (hekousion) at least in the Nicomachean Ethics connects only the ‘up to us’ with moral responsibility and confines the ‘up to us’ to what has been sanctioned by rational proairesis.57 Against this I think Aristotle implies that all that is voluntary is also up to us. He sometimes includes this requirement directly in the definition of voluntariness58 and sometimes makes it an implication of the demand which is itself included in the definition of voluntariness for an internal origin of action.59 Nor is it unreasonable of Aristotle to suppose that despite lacking proairesis the dog which bites you can be blamed. Animals Were held morally responsible by other philosophers too possibly by Democritus60 certainly by Clodius or Heracleides of Pontus61 and tame animals by Epicurus.62
Alexander and Aquinas: proairesis ubiquitous in all action that is up to us
It is rather I believe the much later Aristotelian Alexander who makes the moves just described. In speaking of what is up to us he ties it unlike Aristotle to proairesis. It is found only in beings capable of proairesis and (I think he means) only when they are using their proairesis. In this he conforms at least verbally to the Stoic Epictetus who had said that nothing is up to us except what falls under proairesis in his rather different sense of the term.63 Alexander is motivated to show that Aristotelianism can match the intellectualist presuppositions of Stoicism. He not only accepts the Stoic linkage of what is up to us with proairesis but borrows the Stoics’ own terminology in linking proairesis with rational impulse (logikē hormē).64
Thomas Aquinas compromises but is closer to Alexander's intellectualist account in his discussion of proairesis which he translates into Latin as electio.65 He claims that voluntariness in the primary sense does depend on electio so that animal behaviour is voluntary only in a secondary sense.66 Similarly Alexander in a treatise available to Thomas had reported that some Stoics allowed a weaker sense of ‘up to it’ to be applicable to the doings of animals.67
Alexander's new move has the significant effect of making proairesis ubiquitous in all action that is up to us. This gives it something in common with concepts of the will in Descartes and in modern philosophy as involved in every intentional action.68
In another passage Alexander switches attention from proairesis to boulēsis and gives it a special role in preserving our freedom. Impulse and desire (hormē orexis) he says going along with the Stoics’ intellectualist account are cases of assenting that something is choiceworthy (epi tisi sunkatathesis hōs hairetois). But he warns in opposition to Stoic determinism hormē will not necessarily lead to action if boulēsis does not concur (sundramein).69
The Stoics: will related to voluntariness but not distinct from reason
The Stoics I believe come closer than Aristotle to a full-blooded idea of the will but there are still some very important differences. One difference was noticed in Chapter 2. When Seneca describes anger as involving an act of will (voluntas) to the effect that (tamquam) we should be avenged70 he is using voluntas in a broad sense to refer to impulse. What is significant is that he does not contrast will as a type of conation with cognition. He intellectualizes it treating it as merely one type of cognition: assent to a proposition about how it is appropriate to react. This fits perfectly with the view ascribed to the Stoics in general that impulse (hormē) is assent to a proposition71 and in particular is assent to the appearance that it is appropriate (kathēkei) to act.72 Assent to appearance we know is a judgement. Impulse is also described intellectualistically as reason (logos) commanding (prostaktikos) us to act.73 It matters that the command is said to come from reason. I dissented from the view that Seneca innovates and dissociates will from intellect.74
It is important to see that the Stoics are going beyond Plato's Socrates in their intellectualism. I argued in Chapters 1 and 20 that Socrates’ few restrictions in Plato's early dialogues on what beliefs are compatible with wanting fall far short of Chrysippus’ bold idea that wanting simply is the intellectual judgement that a certain act is appropriate. I believe that will has traditionally been thought of as much more distinct from rational judgement than that.
Although Seneca treats wanting here as an intellectual judgement in the same paragraph he takes a step in the direction of a fuller concept of the will by connecting the will (voluntas) with the notion of voluntariness (voluntarius) and hence with moral responsibility. This connection it has been pointed out is one which comes out in Latin but not in Greek75 since the Greek word for voluntariness hekousion has no connection with words for will. The link between the Latin terms is found already in Cicero.76
Seneca is using the word voluntas in a wide sense for any desire or hormē. But sometimes voluntas and boulēsis are used in a narrow sense for an attitude that only the sage achieves77 and then the Stoic usage has a further implication. Since the sage is supposed to be infallible he knows what is really good and so his will must be a desire not merely for the apparent good but simply for what is good. A non-sage could presumably achieve a similar result by desiring what he believes to be good only with the reservation discussed in Chapter 2 ‘if God wills’. He can then set his heart on what he believes to be good only in so far as it actually is so. But a non-sage's desire will be called a boulēsis or voluntas only in a looser sense.
The Stoics: interrogation of appearances versus Posidonius’ willpower
Although there were developments in the direction of a fuller concept of will for most Stoics a big gap remains: no notion of will-power is at all prominent because their account is more intellectualist. Great moral effort is required but as seen in Chapter 15 it is the intellectual effort of questioning appearances. We saw Epictetus in a passage translated there telling his students to practise questioning the appearance that the beautiful or grand passer-by involves something good or the bereaved or hungry person has encountered something bad.78 Admittedly Epictetus does refer to the questioning of appearances as a struggle (agōnisteon)79 but the process is described in intellectual terms.
Other references to strength or domination are also intellectualized by the Stoics. They do talk of strength and weakness connecting it with tension in the soul which enables it to endure.80 But the weakness is the intellectual weakness of a weak assent: if you have not sufficiently questioned appearances you will have a weak and changeable opinion about what is good or bad.81 There is a similar intellectualizing when the Stoics talk of the ruling (hēgemonikon) or dominating (kratoun kurieuon) part of the soul for it is standardly referred to as reason not as will. And similarly when the runner's momentum is greater than (pleonazei) his impulse to stop we have to recall that the impulse thus overpowered is an intellectual judgement.
There is I believe at least one exception to this lack of reference to will-power. But it is found in Posidonius the Stoic who deliberately reverted to Plato's tripartite psychology. Posidonius takes up Chrysippus’ concession that people sometimes weep without willing to (mē boulomenoi). But he makes an entirely different use of the notion of will. He says that the emotional movements press so hard (sphodra enkeisthai: a military metaphor) that they cannot be mastered (krateisthai) by the will (boulēsis).82 Here the will is turned into something that tries (but fails) to exert power. Similarly some people stop weeping in spite of willing (boulesthai) to continue because the emotional movements can no longer be aroused (epegeiresthai) by the will. The will had not been treated in this dynamic way by Chrysippus. On his account people who weep or cease weeping against their will are receiving conflicting appearances83 presumably appearances about whether things are bad. Theirs is a state of intellectual confusion not a failure of will-power. It is Posidonius’ Platonist sympathies which bring will-power into the Stoic account.
Elsewhere Posidonius tends to speak in terms of reason (logismos) rather than will (boulēsis) but as noted above at least he puts a premium on reason as opposed to high spirit and appetite by following Plato's comparison of it to the charioteer who ought to take control.84
Even Posidonius’ teacher Panaetius had moved a little way in this Platonic direction when he suggested that there is a force (vis) called impulse (hormē) in the appetite (appetitus) and another force in reason (ratio) and that reason presides (praesit) while appetite submits (obtemperet). At least in the temperate person the impulses are obedient (oboedientes) to reason. But the force in reason is not spoken of in this brief citation as a kind of will.85
Epictetus: proairesis connected with freedom and responsibility
Epictetus I have said does not follow Posidonius in speaking in terms of will-power. But he gives renewed prominence to Aristotle's term proairesis.86 Zeno had continued to use this term rather in Aristotle's way87 but Epictetus changes its meaning in a way that brings it much closer to an idea of will which is how I have translated it in Chapter 16 in the passage in which he says:
I will fetter you’. ‘What did you say man? Fetter me? You will fetter my leg but my will (proairesis) not even Zeus can conquer.’
Here and repeatedly elsewhere Epictetus is insisting that my proairesis is free from all constraint.88
Epictetus connects proairesis not only with freedom but also with what is up to us (ephhēmin). All that falls under proairesis is up to us.89 Moreover unlike Aristotle he holds that nothing is up to us except what falls under our proairesis.90 I have already suggested that this may have helped to motivate a parallel shift in Alexander.91 The result is that only the mental is up to us. Epictetus specifies following his teacher Musonius Rufus that the evaluation of appearances is up to us92 and so is assent to those appearances93 and hence the shaping of our proairesis but nothing else.
Had any earlier Stoic anticipated the idea that only the mental is up to us? Antipater head of the Stoic school from about 152 to 129 BC has been named as a possible candidate.94 He described the goal of life as doing everything in one's power (kathhauton) to achieve the natural objectives.95 So it is probably he who made the comparison with an archer and said that the goal is not hitting the target but doing everything one can (facere omnia quae possit) to aim or align (collineare) the arrow right.96 Evidently hitting the target is not thought of as or as necessarily in one's power. What is or is necessarily in one's power is steps towards aiming right. But are these steps sometimes or always mental rather than physical? For all that we have been told physical steps may be at least often in one's power and it has not even been excluded that in favourable circumstances hitting the target may be in one's power. So Epictetus is the first to make it clear that physical activity is never up to us on the grounds that it always could be frustrated.
So far I have said that Epictetus connects proairesis not exactly with will-power but with freedom and with what is up to us. I assume the last means that he connects it with moral responsibility that is with what you can be praised or blamed for. But this inference has been challenged.97 It would mean that someone could not be blamed directly for a physical activity but only for the mental attitudes involved in a physical activity. Can we be sure that Epictetus intended this? For we have evidence that at least some late Stoics broke the connection between ‘up to’ and moral responsibility by allowing that animals’ behaviour is up to them without however holding them morally responsible.98 But I am persuaded99 that Epictetus does not break the connection. For he is prepared to confine moral responsibility as narrowly as what is up to us that is confine it to mental attitudes. We should praise or blame (epainein psegein) people only for their judgements (dogmata) not for indifferents (koina)100 and you are accountable (hupeuthunos) only for the only thing that is up to you (epi soi) and that is the proper evaluation of appearances.101
Epicureans: freedom more important than will
The Stoics’ Epicurean rivals were interested in freedom but not so interested in the will. As others have pointed out102 Epicurus made a very early use of the metaphor of freedom. The text as emended by Usener says of the wise man:
He laughs down that fate which is introduced by some as mistress (despotis) of all and says instead that some things happen of necessity some by chance and some things are because of us (parhēmas). For he sees that necessity is unaccountable and chance unstable but what is because of us has no master (adespoton) and it is to this last that blameworthiness and the reverse naturally belong.103
In fact Epicurus’ usage goes back still earlier to Plato as we have seen.104 But what is missing from the passage is any reference to will. Lucretius takes Epicureanism further.105 He is the first to introduce the expression ‘free will’ (libera voluntas) though he is shortly followed by Cicero who complains that the Stoics preclude free will.106 Lucretius bases its possibility on the unpredictable swerve of atoms. But what seems to do the work in Lucretius’ explanation of freedom is the swerve rather than the will. He is perfectly happy to say that the mind (animus) when it wills (velit) strikes the force of the soul rather than talking of the will acting.107
Perverted will pride and fall: Pythagoreans and Plotinus
I have mentioned Plotinus already as developing Plato's treatment of the choice of one's next incarnation and as using thelēma as a word for the will. But much more important were his views on pride and will as the beginning of evil. For souls that turn away break loose and become ignorant of the Father the beginning of the evil is pride (tolma) and willing (boulēthēnai) to belong to themselves alone. They are pleased with their own self-determination (autexousion) and create the greatest possible distance (apostasis) from the Father.108 The same happens at the level of intellect when it becomes multiple by willing (thelein) to possess everything.109 There is a restless nature originally at rest in eternity which however wills (boulesthai) to govern itself and belong to itself and chooses (helesthai) to seek more than the (timeless) present. This results in the creation of time out of timeless eternity.110
Tolma or pride had played a role in earlier sources too in the creation of lower levels of reality. Thus the Neo-Pythagoreans’ Dyad which provided a model for Plotinus’ intellect was called by them tolma because it separated itself from their version of the One.111 And Irenaeus reports that among certain Gnostics it is tolma which leads to the creation of the physical world.112 But in Plotinus the connection of tolma with will makes closer the relation to Augustine who alludes to Plotinus’ treatise.113
Augustine's clustering
Augustine's treatment of the will is new in more than one way. Most relevantly Augustine brings together all the criteria which we have seen occurring separately in others. Let me illustrate this for each in turn. First will (voluntas) belongs to the rational soul:
To the irrational soul also He gave memory sense appetite to the rational he gave in addition intellect intelligence and will.114
Secondly Augustine connects the will with freedom for the choice (arbitrium) that the will makes is free and one of his best-known treatises is called On Free Choice of the Will (De libero arbitrio voluntatis).
Thirdly Augustine connects the will (voluntas) with responsibility as in the following passage which also talks of perverted will (perversa) a will turned away from God not merely like the reason of Zeno and Chrysippus discussed in Chapter 3 above from one's own better judgement.
It makes a difference what a person's will (voluntas) is like. If it is perverted (perversa) these movements [sc. appetite fear joy grief] will be perverted in him. If it is upright (recta) they will be not just blameless but praiseworthy. Indeed the will is present in all these movements. Rather they are all nothing other than acts of will (voluntates).115
I shall return to this passage in Chapter 26. Another relevant passage connects free choice of the will with responsibility:
And I attended in order to understand what I heard that free choice of the will is the cause of our doing wrong.116
Fourthly Augustine repeatedly speaks in terms of will-power and the failure of will-power. He sees his will as struggling against lust and this will be the subject of Chapter 26. Often he speaks in terms of the will's command and one of the innovations that will emerge in the later chapter is Julian's insistence against him that he should also recognize the will's different role of consent.117
Fifthly Augustine comes to make willing ubiquitous in all action:
Yet if we attend more subtly even (etiam) what anyone is compelled to do unwillingly (invitus) he does by his will if he does it. It is because he would prefer something else that he is said to do it unwillingly (invitus) that is wanting not to (nolens). He is compelled to act by some evil and he does what he is compelled to do through willing to avoid or remove from himself the evil. For suppose his will is so great that he prefers not doing this to not suffering that. Then indubitably he will resist the compulsion and not do it. Hence if he does it it is not indeed with his full (plena) and free will. But because the effect follows his will we cannot say control over his act was missing.118
Augustine reached this position gradually. First he suggests in Confessions 7 that whatever we really do we do by will but this leaves out the evils that one does unwillingly (invitus) because these one may undergo (pati) rather than doing.119 But this exception is put in doubt in Confessions 8 by the view that one has two wills neither of them complete (tota)120 which suggests that reluctant misdeeds may be following the will even if not the complete will. This is confirmed in our new passage among others.121 Unwilling acts follow the will even if not the full (plena) will. That is why Augustine says even (etiam) unwilling acts are done by will. A fortiori all other acts are so done.
Sixthly Augustine develops the criterion of a perverted or bad will. In the City of God122 he quotes a version of Ecclesiasticus:
The beginning of all sin is pride (superbia).
He connects this with the will saying:
What could be the origin of evil will (mala voluntas) except pride?
And he applies this to the Fall of Man saying that the effect on the will of being too pleased with oneself and falling away from God instead of loving him was what made Eve believe the serpent and Adam heed his wife rather than obey God's command. Another fall that of the fallen angels is treated by Augustine123 as it had been earlier by Evagrius124 as due to pride (huperēphania). And Plotinus had seen tolma as causing a descent. Augustine himself applies the message to his own case. Lust was the result and punishment of his own pride when he failed to listen to God.125
A further great innovation of Augustine's is enormously to expand the functions of the will. In On the Trinity for example will performs sonic of the functions of directing attention. It unites perception with the perceptible126 memory with internal vision127 and intellect with objects taken from memory.128 It is responsible for imagination.129 Faith is also due to will.130 Belief depends on the assent of the will.131 We have seen that emotions are acts of will132 and we shall see that will is the centrepiece of Augustine's objections to lust which will be the subject of Chapter 26. The expansion of functions gave the will a greater importance than ever before.
We have seen how the different functions we have discussed gradually got associated in clusters with each other and with a rational desire for the good distinct from reason itself. We can see that Augustine made the most decisive difference. But the associations had started long before him and it will not matter if we talk of concepts of will in earlier philosophers provided we see as we now can how they fall short of Augustine's.
Maximus’ thelēsis and Stoic oikeiōsis
I do not think the concept of the will had to wait until Maximus the Confessor in the seventh century added his contribution. So much had already been brought together by Augustine and what Maximus added did not remain an uncontroversial piece of orthodoxy. I also want to suggest that Maximus’ contribution was not so novel either but was a borrowing from something that had gone before but something from a completely different direction: the Stoic theory of oikeiōsis.
Maximus was defending the view that Christ had two wills one human one divine. But he wanted to explain why Christ's human will could not sin. So he distinguished Christ's human will as a natural will (thelēma phusikon) different from our gnomic will since the latter can turn in either direction towards good or bad according to our opinion. This to scholastics came to seem the right view and Maximus has been praised for defining the natural will as a faculty directed of its essence to the good rather than as something one calls ‘will’ when it happens to be so directed. Another point considered important is that the will aims at this good quite independently of reason although reason recognizes the same good.133 The last point however is not a universally agreed feature of the will since after 1270 it became a matter of debate whether and in what sense the will was independent of reason.134 As for the first point the idea of a naturally directed desire for the good does not seem particularly new. Even before the Stoics Aristotle already holds that everybody naturally desires a happy life.135
In fact it is the Stoics from whom Maximus’ favoured definition of the will seems to derive. No less than five features of the definition he cites (and silently presupposes) proclaim this link. First the good aimed at is self-preservation. Secondly what is to be preserved is described by the Stoics’ word sustasis our ‘constitution’. Thirdly will is said to depend only on nature unlike proairesis. Fourthly the Stoic term sunektikē sunekhein is used when it is said that will holds the substance together. Even more characteristically what it holds together is the idiōmata the attributes which the Stoics postulated as lasting through an individual's life and distinguishing it from all other individuals.
My suggestion is that Maximus’ will (thelēsis) is a variant of the Stoics’ oikeiōsis that attachment that is felt by newborn infants and animals to their own physical constitution (sustasis) and which the adult human can later extend to his entire rational constitution. This attachment drives infants and animals to preserve that constitution. The claim that it is natural is important to the Stoics because they argue against opponents who want to ascribe reason to animals that this penchant for self-preservation is due to nature not to reason.136 The account of the will that Maximus turns out to favour and for which he has been so much praised runs as follows:
They say that natural thelēsis or thelēma is a capacity desirous (orektikē) of what is in accordance with nature a capacity which holds together in being (sunektikē) all the distinctive attributes (idiōmata) which belong essentially to a being's nature. The substance being naturally held together by this desires (oregetai) being and living and moving in accordance with perception and intellect striving for (ephiesthai) its own natural and complete existence (ontotēs). A thing's nature has a will (thelētikē) for itself and for all that is let to create its constitution (sustasis) and it is suspended in a desiderative way over the rational structure of its being the structure in accordance with which it exists and has come into being. That is why others in defining this natural thelēma say that it is a rational and vital desire (orexis) whereas proairesis is a desire based on deliberation for things that are up to us. So thelēsis is not proairesis if thelēsis is a simple rational and vital desire whereas proairesis is a coming together of desire deliberation and judgement. For it is after first desiring that we deliberate and after having deliberated that we judge and after having judged that we deliberately choose (proaireisthai) what has been shown by judgement better in preference to the worse. And thelēsis depends only on what is natural proairesis on what is up to us and capable of being brought about through us.137
The idea of the will as a desire for self-preservation continues in the fourteenth century.138 My suggestion is that this coming through Maximus may be a Stoic legacy.139 I shall not discuss the immediately following lines of Maximus’ text in which he describes the stages by which will is converted into action because although I believe these stages are of greater interest they are not the ground on which he has been presented as inventing the concept of the will.
I have ascribed to Augustine the originality of bringing all the criteria together. But it is a different question whether bringing them together is a good idea. I believe new reasons would need to be found and indeed a recent work has offered a rationale to show that some such clustering round a concept of will is required in order to show what human action involves.140 But without a new rationale we have little incentive to accept the clustering.
The idea of perverted will involves a metaphysics that is not now widely shared. As for the idea of will as ubiquitously present in all action people may nowadays be more sympathetic to Aristotle's idea that in voluntary action what is always present is an internal cause141 always I think desire sometimes negligence in addition142 but not always rational will. Aristotle's examples include such baser desires as anger143 and appetite.144
The idea of freedom may be better treated without the idea of will. Certainly Lucretius’ idea of an uncaused swerve in the operation of the will seems to me unhelpful leaving us caught in the dilemma as to whether our actions are inexplicable or necessitated. I have sought elsewhere to tackle this dilemma by arguing that actions can be fully explained and indeed caused without being necessitated.145 But there are other treatments of freedom too which feel no need to invoke the idea of will and I shall mention another shortly.
As for moral responsibility Aristotle's view is persuasive that it extends wider than just to actions and agents motivated by rational will.
For some idea of will-power there is a good use. We need to describe the effort to pursue what we think best against desires of which we approve less. But this phenomenon may be better analysed in the way just indicated in terms of different layers of attitudes. Desires at the first level may be the subject of second-order approval or disapproval. The effort to act in accordance with the approved desires requires what we call will-power. But I doubt that anything is gained by thinking of this effort in terms of the exercise of a rational faculty rather than in terms of the varied thoughts imaginings and acts of attention involved.
One element in the notion of freedom has also been analysed in recent years in terms of first-and second-order attitudes. Freedom involves being able to give second-order approval to one's attitudes of the first order.146 To this extent there may be some overlap between treatments of freedom and of will-power. But on the whole we may think it was more reasonable of Plato Posidonius and Galen to handle them quite separately.