William King on election, reason, and desire: a reply to Kenneth Pearce

ABSTRACT William King’s De Origine Mali has recently started to attract some attention in early modern scholarship. In a recent paper devoted to King’s theory of free will, Kenneth Pearce identifies a “lacuna” in his text, namely the fact that King “never explicitly describes the process whereby election leads to action” (Pearce, “William King on Free Will”, 4). In this paper, I analyse King’s theory of ‘election’ (roughly, free choice) and Pearce’s interpretation of it. I discuss his claim that there is a lacuna in King’s account and argue that the text provides us with important suggestions on how election generates action. Therefore, no speculative proposal to fill the lacuna is needed. This, in turn, allows me to develop a reading of King’s text that avoids a textual puzzle unsolved in Pearce’s interpretation: while he maintained that every election is “with reason”, my account can take King’s text at face value and explain how some elections are reasonable while others are not.


I. Introduction
Although largely forgotten by present-day philosophers, Anglican divine William King (1650-1729) was well known in his time. 1 King was Archbishop of Dublin from 1703 to 1729 and an important figure in late seventeenth-and early eighteenth-century European thought.His ideas were influential on some of the most prominent thinkers of the time, including John Locke (see James Harris, Of Liberty and Necessity, 42-46; Storrie, "William King's Influence"), Francis Hutcheson (Boeker, "Francis Hutcheson on Liberty"), George Berkeley (West & Fasko, "The Irish Context"), and David Hume (Fisette,  "Hume's Quietism", 78-81; Mossner, "Hume's Early Memoranda", 496).His main philosophical text, a volume entitled De Origine Mali, immediately attracted a great deal of attention upon its publication. 2It was famously discussed by both Bayle and Leibniz, and was soon translated it into English. 3ing's book includes an original account of the faculty of free will (called 'electio' in the Latin) and constitutes a sophisticated contribution to the contemporary debate on this subject.Despite all this, early modern philosophy scholars rarely devoted attention to it. 4n a recent paper entitled "William King on Free Will", Kenneth Pearce has provided the first full-fledged scholarly interpretation of William King's theory of election.By doing so, Pearce has successfully shown that "King's account of free will is in fact more interesting, original, and sophisticated than has previously been recognized" (Pearce, "William King", 1).My aim in this contribution is to build on Pearce's work while also contesting two problematic features of his interpretation.The paper is structured as follows.In Section II, I retrace Pearce's interpretation of King's theory of election and his claim that for King every act of election is "with reason".In Section III, I explore Pearce's claim that there is a "lacuna" in King's account between election and action, I analyse and criticize his speculative filling-in, and argue that King's text provides us with important suggestions on this point.Therefore, no speculative proposal is needed.In Section IV, I show that an important part of Pearce's interpretation of King's account of election, which he developed in order to build his solution to the supposed lacuna, departed from the plain sense of King's text.Once we recognize that there is no such a lacuna in King's philosophy, a more natural interpretation of the text is possible.Finally, in Section V, I show how my reading allows us to take what King says about the relation between reason and freedom of the will at face value.

II. Pearce's interpretation of King's theory of election
Pearce's paper has the merit to show with great clarity King's dissatisfaction with both sides of his contemporary debate on freedom of the will which opposed two factions: "necessitarians" and "libertarians" (Pearce, "William King", 2).Pearce explains that, on the necessitarian view, as King conceives of it, there is simply no liberty of the will: one is free only when one's actions are not constrained by external obstacles.On the libertarian view, King believes, the liberty of the will is a faculty that can only be used for ill, as we would be better off were we always determined to choose the greater good.In other words, libertarians do not explain why it is better for us to have free will rather than simply be determined to act according to the greater good.
Pearce's convincingly shows that King's original position, developed as a solution to this debate, consists of arguing that, by an act of election, human beings can make the option that they choose better than it was before the election.In other words, free will is a faculty that is good for human beings to have because it generates value.King's position, therefore, is close to libertarian views, but does not have the problematic conclusion that we would be better off without free will.Because of the fact that we have free will, King believes, we can elect one option among others and, by electing it, we can make it better than other competing options, all other things being equal.
King's position on free will is not only original in the context of his contemporary debate, but also interesting as an early account of a form of what we now call constructivism about value.Human beings are creatures who, at least in part, generate value by choosing one course of action over another.Despite recognizing the interesting nature of King's philosophical work on the constitution of value, Pearce laments the presence of a "lacuna" in his account, namely the fact that "King never explicitly describes the process whereby election leads to action" (Pearce, "William King", 4).According to Pearce, this lacuna is so important that "is in large part responsible for the misunderstandings and dismissals of King's account" (4).Pearce's attempt to make "some progress … on this issue" lead him to provide a "somewhat speculative suggestion" (5) on how this process works.In order to understand the suggestion (fully developed in Section III), it is important that we summarize his interpretation of King's account of election, as presented by Pearce in three main points: (i) First, the act of election determines the understanding's judgment about which action is best, rather than the other way around.
(Pearce, "William King", 5) King clearly stresses that election generates value: "this goodness of the object does not precede the act of election, so as to elicit it, but election brings about the goodness in the object, i.e. the thing pleases because it is elected, it is not truly elected because it pleases" (5.1.3.17,113, Pearce's translation).And again: "things please this agent [with the faculty of election], not because they are good in themselves (in se bonae) but are made good ( fieri bonas) because they are elected (eliguntur) " (5.1.4.1, 117).As Pearce puts it, King's original proposal is that "free beings possess a faculty of election which is successfully exercised whenever it gets what it elects" "William King", 4. It is for this reason that the election of one option over another determines, other things being equal, the understanding's judgement.Presented with two equally good options, the understanding does not judge one of them to be better nor does it prefer one of them.
After the election, however, one of the two options, the one elected, acquires new additional value: the understanding, therefore, judges that option as being better.
The second main point of King's theory, as Pearce reconstructs it, is that the election which determines the understanding does so because it alters the value of the options among which the agent can choose: (ii) Second, this [the fact that the election determines the understanding] occurs because the act of election alters the values of objects.
(Pearce, "William King", 5) As Pearce explains, King's theory is not merely a theory of liberty as indifference.It is not simply the case that we can choose freely between two options, regardless of their value.We can do that, but by choosing one option rather than the other, we also confer some value on the elected option and, therefore, one of the two initially neutral options becomes (more) valuable.We freely elect one rather than the other, and those options which were neutral to us before the choice become valuable in different ways.
Pearce usefully proposes to clarify this point by applying King's theory to the case of Buridan's Ass, a donkey that is placed at the same distance from two different piles of hay and, having no reason to prefer one over the other, starves to death.If the donkey had the faculty of election, it could bring it about that the two piles are no longer equally attractive.This can be done without any change in the intrinsic features of the piles or the donkey's beliefs about those intrinsic features: All that is needed is a change in the donkey's values, a preference for e.g. the left over the right.This preference or valuation is arbitrary, but the resulting action is not, for the preference itself confers additional value on the left pile which removes the indifference and permits rational action.(Pearce, "William King", 4)   According to Pearce, "rational action" follows from the election because the choice augments the value of one of the two options.Finally, Pearce believes that King's view is not vulnerable to a charge of arbitrariness or acting by mere chance, because the act of election happens "with reason".In the third phase of his reconstruction, he affirms: (iii) Third, because election alters the values of objects so that the object chosen is best, the agent can be said to elect 'with reason' and the agent's action can be said to be determined by reason rather than chance.
(Pearce, "William King", 5) Pearce reads King as arguing that "free actions" are "undertaken for reasons" (5).The idea here is that King wants to "argue that action is in some sense determined by reason and not chance" but also to say that "elections … are clearly not determined by reason" (5).In other words, Pearce stresses the fact that there is a distinction between election and action.On the one hand, election is not determined by reason, otherwise the first point of King's theory about the priority of election would be invalidated.On the other hand, actions that follow from election can be said to be determined by reason: "King's view must be that the act of election allows for the action to be determined by reason" (5).
To support the third phase on the reconstruction of King's account of election, Pearce relies on the following passage, which I will call the Reason Passage: Reason Passage: "You will say: If these things are true, this agent will be determined to act by chance not reason.… As for reason, whoever puts a lesser good ahead of a greater [good], must be judged to have elected without reason, but one who by electing makes that a greater good, which prior to election had no good or less good in it, that person certainly elects with reason" (5.1.3.18,113-14, Pearce's translation) According to Pearce, by electing one of the options the agent "creates a reason that wasn't there before and this reason in turn determines the agent to action" (5).This is why, in Pearce's view, King says that an agent can elect with reason, even if "elections … are clearly not determined by reason" (5).The action that follows from the election is determined by reason, while the election itself is not.In this explanation, "King's use of the phrase 'elects with reason' (cum ratione eligit)" remains for Pearce "admittedly a bit puzzling" (5).

III. The alleged lacuna between election and action
As I indicated in the previous section, in the context of his reconstruction of Kings theory of election, Pearce laments the presence of a "lacuna" in King's account: the fact that "King never explicitly describes the process whereby election leads to action" ("William King", 4).
To fill this gap between election and action, Pearce proposes an "admittedly somewhat speculative suggestion" according to which "what King sometimes calls 'determination to action' is in fact a second act of the will distinct from election, and this act is determined by the judgment of the understanding" (Pearce, "William King", 5).In other words, Pearce proposes to read King as affirming that, first, free acts of election produce value and, second, that the value generated in this way determines the agent to act.This determination to action is also an act of election ("a second act of the will", 5), but one that can be said to occur with some reason (cum ratione) rather than by way of undetermined election.
This interesting attempt to clarify the relation between election and action, however, is problematic for two reasons.First, it is not supported by textual evidence.While Pearce provides a reading of the reason passage that allows him to construct his proposal, the passage remains "puzzling" even by his admission.A second and more important concern is that Pearce's suggestion is liable to a regress problem.On the one hand, if the determination to action is determined "by the judgment of the understanding", it is unclear in which sense it is an act of the will, since for King the will is a free faculty of election. 5f, on the other hand, the action is not determined by the understanding, the alleged "second act of the will" would have to be a second act of free election, with an evident regression problem.In other words, either the determination of the mind following from the election is not free and, therefore, is an act of the will; or it is free, but we then have a second election which, in turn, will need to determine the action, and so on.Because of the lack of textual evidence and this conceptual problem, Pearce's suggestion appears unsatisfactory.
While I agree with Pearce that King does not say enough on the connection between election and action, he does say something that is both interesting and relevant to this discussion.Rather than developing an alternative speculative account, then, I want to suggest that we need a closer look at King's text.With some extra passages in view, it will emerge that lacuna identified by Pearce is only apparent.
The first indication of the relation between election and action in King's account appears in 5.1.3,the subsection in which he introduces his theory of election.Here, he clarifies that "the agent endowed with it [the faculty of election] cannot be determined in its operations by any goodness preexistent in the object (bonitate praeexistente)", rather, "the agreeableness arises from the determination (conventientia … ex determinatione oriri)" (5.1.3.5, 107).Right after this, King introduces a conative element that is not discussed by Pearce.Let us call the following excerpt the Desire Passage: Desire Passage: "Let us suppose this power [of election] to be already determined (it does not matter how) to embrace a certain object, or to exert the proper operations relating to it; it is certain that desire follows this determination (desiderium sequi determinationem), and desire is followed by an endeavor to acquire and enjoy the object (desiderium vero conatum acquirendi & fruendi objecti) following the application of the power.But if anything should hinder or impede this endeavor (hunc conatum) thereby preventing the power from exerting those operations which it undertook to discharge in relation to the object, then indeed uneasiness would arise from the hindrance of the power." ( 5.1.3.6, 108)  In this passage, King introduces desire as the result of the determination of an act of election.King explains that the act of election creates a "determination of the power itself" (determinationem ipsius potentioe, 5.1.3.6, 108) and determination is followed by a desire "to obtain the object" that has been elected. 6 few paragraphs later, King specifies that the very determination is the election, saying that "we shall call this determination an election" (Determinationem hanc electionem dicemus, 5.1.3.16,113). 7e can further complement this reconstruction of King's account with what he writes in 5.1.5.23: the mind judges things to be good because we have willed them, because we have formed an appetite in ourselves by some antecedent election (quia antecedente aliquâ electione appetitus nobis creavimus), and those things that we embrace by this factitious appetite (per factitium hunc appetitum), as we may call it, give us no less pleasure (non minus placent) than that which we desire by the necessity of nature.
(5.1.5.23, 139) Here King affirms that by election, we are able to form an appetite which is not natural but artificial or factitious.This appetite is a consequence of the election and is therefore brought about by free human choice rather than simply by nature.That appetite, in turn, generates an endeavour to action.
Given that this factitious appetite plays the same role that in the earlier passage King attributed to desire and given that they are both conative elements, my hypothesis is that they are considered equivalent by King.A free election determines itself and, as a consequence, generates a "factitious" conative element that manifests into action.The supposed lacuna in King's discussion of election and action is filled, with no need of speculation.As I show next, once the lacuna has been filled in this way, we can make sense of King's claim that some elections are reasonable even if they are not determined by reason, which remained a puzzling fact in Pearce's interpretation.

IV. Election with reason and reason for action
Once we have rid ourselves of the need to fill in the supposed lacuna we can provide an interpretation of the Reason Passage that, unlike Pearce's, does not depart from the plain sense of the text.While Pearce's original paper illuminates many interesting aspects of King's theory of election, in his interpretation of the Reason Passage it remained unclear (more than "a bit puzzling") why King affirms that one can elect with reason rather than act with reason, if, as Pearce maintains, the reason is only a reason for action rather than a reason for the election.Recall, the third point in Pearce's reconstruction of how election works was to say that "because election alters the values of objects so that the object chosen is best, the agent can be said to elect 'with reason' and the agent's action can be said to be determined by reason rather than chance" (Pearce, "William King", 5).Pearce's interpretation, therefore, must maintain that the Reason Passage contains a mistake, or at least an oversight, on King's part, as he talks of election rather than action being cum ratione.
An additional related problem, for Pearce's view, is that it seems that King must maintain that every election is cum ratione.He writes: "[t]he agent can be said to elect 'with reason' insofar as the agent's election creates a reason that wasn't there before and this reason in turn determines the agent to action" (5).However, this is problematic.As Pearce recognizes, King believes that every election gives one a reason for the action that would promote the object of that election.If Pearce is committed to what he affirms in the quotation above, therefore, it follows that every election is cum ratione.But in the Reason Passage, King clearly says that "whoever prefers a lesser good to a greater one must be judged as electing without reason" (qui minus bonum majori praetulerit, sine ratione eligere censendus est, 5. 1.3.18,114). 8Contrary to what Pearce suggests, therefore, it seems that one can elect with or without reason.
King's Reason Passage is admittedly ambiguous.We could read him as saying that an election is "with reason" anytime it makes a good greater than it was before.If this were the case, we would have to admit that every election is cum ratione, as Pearce does.This is so because, for King, any election generates some value that is added to the elected object by virtue of its being elected.But this is not the only available reading of the text.Another, more charitable reading, sees the text as saying that election cum ratione is possible when it does not fall into the category of unreasonable elections mentioned in the previous sentence.In other words, an election that chooses a lesser good over a greater one is without reason unless the election makes the elected good greater than the other.This reading is more charitable than Pearce's because it does not imply that every election is cum ratione.For Pearce "[t]he agent can be said to elect 'with reason' insofar as the agent's election creates a reason that wasn't there before" (Pearce, "William King", 5).But this is how every act of election seems to work according to King, who explicitly affirms that the mind by an act of election generates a reason (ratio).When we elect, we acquire a reason to pursue what is elected.As King says: this reason is made by the mind itself (ratio haec ab ipsa mente ficta) and may serve for every election equally, since it is drawn from the indifference of the will itself and he who does anything for a reason which he himself made (propter rationem a se fictam) and is indifferent to either side, must be considered as if he had acted without any reason (ac si sine omni ratione egisset).
(5.1.5.24, 139) King is saying that by electing one option over others, the agent creates a reason for him to pursue the object of his election.However, because this reason does not produce the election but rather it is produced by the election, we can say that the agent acted without reason.In other words, any election always generates a reason for acting.The following action can be said to be free rather than determined by reason, because the reason for action is created by a free act of the will.Therefore, Pearce is correct in thinking that every action that derives from election is an action with reason, but not in saying that it is an election with reason.
The problem with Pearce's interpretation, as I see it, derives from the fact that in the attempt to provide a speculative filling-in for the supposed lacuna he identified between election and action, he is forced to depart from a plain reading of King's text.Because he wants to fill the lacuna appealing to the fact that there must be "a second act of the will distinct from election … determined by the judgment of the understanding" (Pearce, "William King", 5), he is forced to affirm that every election is cum ratione.Every election is cum ratione because it produces a reason that in turn determines the understanding and leads to action.However, by using the Desire Passage, we can explain away the need for a speculative filling-in.Consequently, we can take the Reason Passage at face value: some elections are with reason while others are not.

V. Limitations on election
It remains necessary, in this last section, to clarify what makes an election with or without reason.As we saw above, the relevant factor here cannot be the fact that elections produce reasons for action, because that is true of every election, but King affirms that some elections are without reason.
As we saw, according to King, the emergence of a factitious appetite follows and is determined by the free election, and, in turn, it generates an endeavour to action.We are free to control the inception of this process, but we are passive with respect to its consequences once our election is determined.This is important to keep in mind, as it allows us to understand why King stresses the fact that desire or factitious appetite following from election can be frustrated.If the object of these conative reactions cannot be obtained, the agent will naturally suffer: if one elects an option that cannot be realized, "then indeed uneasiness (molestia) would arise from the hindrance of the power (ex potentiâ impeditâ oriretur)" (5.1.3.6,108).
King is clear that, in such cases, the agent would have been better off electing something else.Even if the act of determination makes an initially neutral object more valuable than it was in the eyes of the agent, this does not mean that every elected object becomes the best option for the agent to pursue.King makes this very explicit just a couple of paragraphs after the Desire Passage, where he goes on to detail some limitations for the possibility of election.These are not limitations on the power of electing, which is free by definition.They are rather normative limitations: they limit what the agent ought to elect without removing the possibility of free choice.They are practical constraints that every agent should consider when electing among various options.In what follows, I argue that these are constraints that the agent should consider in her action if she is to be rational.As I will show, it is with respect to these constraints that an election can be with reason or without reason.These limitations are the following.
(L1) The object of one's election must be possible to obtain: he that attempts this [something impossible], must necessarily be unhappy (infoelix) in the event, because, since the thing which the power undertakes cannot be done, uneasiness (molestia) must necessarily follow the frustration and hindrance of its exercise (5.1.3.8, 109)   This limitation, King says, is the only one that applies to both agents with finite and infinite powers.The next two limitations apply only "if the agent's power be finite " (5.1.3.10,110).(L2) The election must not determine anything which exceeds the ability of the agent: the agent … must consult his abilities, and not to determine itself to anything which exceeds them, otherwise it will be no less disappointed in its endeavor (conatu frustrabitur) than if it attempted absolute impossibilities.
( 5.1.3.10,110)   (L3) The election must consider and regard the affects: this agent should have some regard to these [natural] appetites (appetituum), and not disturb them unnecessarily, nor restrain them from appropriate enjoyment of their objects.The person who does this will bring upon himself uneasiness and unnecessary struggle (molestiam & luctam minime necessariam).
( 5.1.3.15,112)  These three limitations appear a few paragraphs after the Desire Passage and a few paragraphs before the Reason Passage. 9In my interpretation, they are essential to understanding what King believes about relation between election, desire, and reason.King himself in one of the final paragraphs of 5.1.3affirms that "not all things are indifferent to this power [of election] and indeed it admits of some limitations (limitationes), as it was observed, beyond which it necessarily makes happiness perish (foeliciatate cadat) " (5.1.3.20,115).The practical constraints analysed in this section should therefore be conceived as limitations that the agent should consider in order to be happy with his elections.It is, presumably, for the same reason that King affirms in the Reason Passage, that "whoever puts a lesser good ahead of a greater [good], must be judged to have elected without reason" (5.1.3.18,114, Pearce's translation).The reason passage, which appears in King's text after the three limitations, seems therefore to be a fourth constraint that could be summarized as follows: (L4) The election ought to be a preference for the greater good unless by the act of election itself an even greater good can be created.
As we have seen above, a desire or appetite follows the determination of any act of election.Anytime we value something by choosing it, we feel a desire to obtain the object of our choice.But King has also affirmed in 5.1.5.24, that every election produces a reason for action.The most charitable and plausible reading is, therefore, that every election produces a reason for action (which, I have proposed, consists of the factitious appetite or desire to acquire the object of election).In this sense, ratione means simply a "consideration in favor of something".This, however, does not mean that every election is rational (cum ratione).Only elections that respect the practical constraints (L1) to (L4) are rational elections.
The problem with Pearce's view is that it overlooks the fact that election produces a desire or factitious appetite.Consequently, he misses the fact that this desire is a reason for action that is produced by the election, but it is not why King thinks that we can elect with reason.In other words, Pearce's interpretation conflates the reasonableness of election with the fact that every election produces a reason for action.While every election produces a desire which is a reason for action (a consideration in favour of something), the election itself is not always reasonable.When the election consists of a choice of what is worse and the election itself does not add 9 Similar limitations are also listed in Chapter 5, Section 3, "Of Undue Elections".There King stresses that elections that do not respect these conditions are undue (or, following Pearce's translation, 'impermissible').I focus on the first appearance of the limitation (in 5.1.3)rather than on the discussion of their moral role in (5.3) because only in 5.1.3are they connected directly to reason and the reasonableness of election, which is the main point of my paper.
enough value to that option to make it better than other options available to the agent, then the election is irrational (sine ratione).An election that violates one or more of the four constraints above is an election without reason.
On this interpretation, we can maintain that King is consistent in affirming that elections can be with or without reason, and yet that acts of election are free and not determined by reason.While Pearce could not explain why King says that we can elect with reason rather than act with reason, my interpretation provides a plausible explanation for this: King thinks that there is a standard of reasonableness for elections that is external and prior to those elections.One's free will remains absolutely free as it is not determined by these external reasons, but one's elections can still be said to be reasonable or unreasonable depending on their conformity to this standard.On my reading, we elect without reason when the election violates one or more of the four practical constraints.If this is correct, it is clear how elections can be not only reasonable but also unreasonable.This is the case in spite of the fact that every election (reasonable or unreasonable) produces a desire or factitious appetite that counts as a reason for action for the agent.In this way, both Pearce's supposed lacuna and the puzzlement in front of the Reason Passage disappear.