THOMAS REID ON PROMISES AND SOCIAL OPERATIONS OF THE HUMAN MIND

: My paper offers a new interpretation of Reid ’ s account of social operations of the human mind. I argue that it is important to acknowledge the counterpart structure of social operations. By this I mean that for Reid every social operation is paired with a counterpart operation. On the view that I ascribe to Reid, at least two intelligent beings take part in a social operation and the social operation does not come into existence until both the social operation and its counterpart operation have been exercised and the relevant mental thoughts made known to the other being by words or signs.


Introduction
My main topic in this paper is Thomas Reid's account of social operations of the human mind. 1 Reid argues that all operations of the human mind can be divided into social and solitary and that this distinction 'has a real foundation in nature' (EIP, I. 8, p. 68). Yet what are social operations of the mind and how does Reid distinguish them from solitary ones? According to Reid, social operations 'necessarily imply social intercourse with some other intelligent being who bears a part in them' (EAP, V.6, p. 330). Examples include asking a question, testifying a fact, giving a command, or making a promise. Social operations must be 'expressed by words or signs, and known to the other party' (EAP, V.6, p. 330). Solitary operations, by contrast, 'may be performed by a man in solitude, without intercourse with any other intelligent being' (EAP, V.6, p. 330). For instance, seeing, hearing, remembering, judging, or reasoning do not presuppose the presence or existence of another intelligent being, nor do they need to be expressed.
It is worth noting that although social operations are significant in Reid's philosophy for explaining why humans are social creatures, he does not restrict human sociability to social operations. Rather Reid acknowledges that in addition to social operations there are social affections and both play an important role in our social lives. 2 Reid sometimes also calls social operations 'social intellectual powers' (EIP, I.8, p. 69) or 'social intellectual operations' (EIP, I.8, p. 69) to distinguish them from social affections. Social operations and social affections have in common that 'both suppose a conviction of the existence of other intelligent beings' (EIP, I.8, p. 69). According to Reid, affections are directed towards persons 'and imply, in their very nature, our being well or ill affected to some person, or, at least, to some animated being' (EAP, III.ii.3, p. 107). Affections do not require a reaction from the person to whom they are directed. As I will explain in more detail in the following, social operations, by contrast, require such a reaction.
For the rest of the paper, Reid's account of social operations will be my main focus. Current Reid scholarship has not paid much attention to it, and I believe that the few existing interpretations can be improved. 3 This is my main aim in this paper. In my view, Reid scholars have not given sufficient consideration to the counterpart structure of social operations. The view that I defend in the following draws attention to the fact that for Reid social operations come in pairs. Each social operation is paired with a counterpart operation. For instance, promising is paired with accepting a promise, commanding with obeying, asking a question with answering, or testifying with testimonial belief. My view is that a promise does not come into existence until the promisor A has expressed a promise to another intelligent being, the promisee B, and B has agreed to accept the promise and made this known to A. By agreeing to accept the promise B exercises the counterpart operation to promising. By the same token, the other social operations do not come into existence until they have been expressed and the other intelligent being who takes part in it has understood it and reacted to it by exercising the counterpart operation and expressing their reaction. This counterpart structure of Reid's social operations has been widely neglected, but I believe that it is central for properly understanding Reid's thinking about social operations.
In the following, I pay close attention to Reid's understanding of promises. There are two reasons for this. First, it is helpful to focus on a specific social operation. As we will see, many of the considerations that apply to promises apply also to other social operations more generally. Second, Reid's account of promises in Essays on the Active Powers of Man is directed against David Hume's view that fidelity to promises is an artificial virtue. 4 Reid criticizes Hume for reducing virtues to solitary acts of mind and for failing to realize that in addition to solitary acts of minds, there are social acts of mind. Thus, focusing on promises will not only shed light on Reid's thinking about promises but also illuminate how he attempts to overcome problems that he identifies for Hume's philosophy.
My paper proceeds as follows. I begin by asking in what sense social operations are 'social' (Section 2). I approach this question by examining the different components of social operations. I then further clarify my interpretation, outline its implications, and address potential objections (Section 3). Next, I show how my interpretation avoids problems that arise for Rebecca Copenhaver's interpretation (Section 4) and how it advances Gideon Yaffe's interpretation (Section 5). Neither Copenhaver nor Yaffe pay proper attention to the counterpart structure of social operations. I end by considering whether Reid's criticism of Hume's view that fidelity to promises is an artificial virtue is fair and assess possible responses that could be made on Hume's behalf (Section 6).

In what sense are social operations 'social'?
I now turn to the question in what sense social operations are 'social'. Before addressing this question, it is helpful to turn for a moment to solitary operations and their constituent elements. In Essays on the Intellectual Powers of Man, Reid distinguishes the content of a mental operationor 'object' as he calls itfrom the mental act itself and the subject of the mental operation. 5 For instance, let us consider the sentence 'I perceive a lemon tree'. In this case, I stands for the subject, namely, a person or mind, the active verb perceive denotes the mental act of perceiving, and the object of this mental operation is a lemon tree. If we follow Reid in distinguishing the constituents of mental operations in this way, it becomes possible to account for the difference between various types of mental operations such as perception, memory, or imagination. Although all these mental operations can have a lemon tree as their object, Reid would argue that perceiving, remembering, and imagining are three different mental activities. 6 With regard to perception, Reid claims that the object of perception must be something present, while the object of memory must be something past. 7 Moreover, he argues that perception differs from imagination, insofar as in the case of perception the subject has a belief or 'full conviction' that the things perceived exist, while there is no such belief if one is imagining. 8 Reid further notes that the object of perception must be something external. Otherwise if the object is something internal and present, he maintains, it is the object of consciousness rather than perception. 9 On this basis, let us turn to social operations. Social operations differ from solitary operations insofar as at least two intelligent beings take part in them, while solitary operations have one subject. Moreover, Reid makes clear that every social operation, in contrast to solitary operations, must be expressed by words or signs. 10 This has led some interpreters to propose that social operations are speech acts for Reid. 11 The difficulty with this proposal is that it cannot easily make sense of Reid's view that social operations are mental acts. 12 Rather than identifying social operations with speech acts, I believe that we should take him at his word and acknowledge that social operations are mental acts, which are always accompanied by an act of expression. 13 On this view, the mental act, say the mental act of promising, is a distinct act from the act of expression, namely, the utterance 'I promise you to φ'.
Taking these differences between solitary and social operations into account, social operations can be said to have the following constituents: First, two or more intelligent beings must take part in the operation. Second, a social operation must have a particular content, which Reid would call its object. For instance, making a promise involves promising something, or giving testimony involves testifying something. Third, a social operation involves a mental act and additionally an accompanying act of expression. Now that we have identified these different constituents of social operations provides more fine-grained resources for examining in what sense social operations are social. I propose that when asking in what sense social operations are social, we should ask which of the constituents, namely, the presence of two (or more) intelligent beings, the object (or content), the mental act, and/or the act of expression, make the operation social.
For present purposes, it will be sufficient to focus on promises. Like every social operation, promises presuppose the existence of at least two intelligent beings, namely, a promisor A and a promisee B. Making a promise involves promising something φ. If A promises B to φ, say A promises B not to drink in the evening, then the promise cannot be social purely in virtue of its content, or 'object' as Reid would say, because A can take other mental attitudes towards the same content, which are solitary acts of mind. For instance, A can believe that they will not drink, A can desire not to drink, or A can be disappointed that they will not drink. 14 Therefore, it is more plausible that promises are social in virtue of the presence of two intelligent beings, namely, the promisor A and the promisee B, and/or the mental act of promising accompanied by the act of expression.
The presence of another intelligent being is not sufficient to make an operation social. For instance, assume that I ask a friend to help me carry furniture on a moving truck. Although I would have struggled to carry the furniture myself, the act of carrying furniture is not a social act, because a strong person is able to do it alone without help of others. 15 Thus, the mere fact that two or more intelligent beings take part in an action does not make the operation social.
Yet Reid is explicit that social operations 'necessarily imply social intercourse with some other intelligent being who bears part in them' (EAP, V.6, p. 330). This intimates that he regards only those operations as social that necessarily require another intelligent being who participates in the act. 16 Let us assume that I play chess with a friend. Would Reid regard the act of playing chess as a social operation? Although playing chess commonly involves two intelligent beings, it does not necessarily do so, because I could also play with a computer. 17 According to Reid, social operations 'suppose understanding and will' (EIP, I.8, p. 68). A computer lacks active power and, thus, lacks a will. Therefore, a computer is not capable of engaging in social operations. 18 It follows that playing chess is not a social operation in Reid's sense. Yaffeand I am in agreement with him on this issueputs the general point well by stating that in order for a mental operation to be social, 'it must imply the existence of other beings who are intelligent and who are exercising their intelligence' (Yaffe, 2007, p. 282).
Despite the importance of the presence of another intelligent being who exercises their intelligence, I believe that an interpretation of Reid's account of social operations remains incomplete without additionally examining whether the relevant mental acts and the accompanying acts of expression, such as the mental act of promising and the utterance 'I promise you to φ', play a role in making the operation social. This question has not received sufficient attention in current Reid scholarship. 19 I want to propose that another intelligent being takes part in a social operation not merely by being present but by reacting to the act of expression and making their response known. More precisely, as Reid states in the following passages, if intelligent being A utters 'I promise you to φ', directed at another intelligent being B, it is important that B understands the act of expression and reacts to it by accepting (or not accepting) what is promised and making this response known to A.
It is obvious that the prestation promised must be understood by both parties. One party engages to do such a thing, another accepts of this engagement. An engagement to do, one does not know what, can neither be made nor accepted. It is no less obvious, that a contract is a voluntary transaction. (EAP, V.6, p. 336) What makes a promise is, that it be expressed to the other party with understanding, and with an intention to become bound, and that it be accepted by him. (EAP, V.6, p. 342) These passages intimate that the mental acts and their accompanying acts of expression, which are involved in social operations, come in pairs. The proposal is that the mental act of promising is paired with the mental act of accepting the promised thing. This same structure extends to other social operations. For instance, asking a question is paired with responding, commanding with obeying, 20 and testifying with testimonial belief.
It is worth examining more closely how Reid understands the pairing or counterpart structure of social mental acts. First, Reid accepts that we can engage in social operations because we have been equipped with original faculties 21 that enable us to engage in social acts such as promising, testifying, asking a question, or commanding. 22 Reid believes that these social acts arise from simple faculties and that they cannot be reduced to other mental acts. In this respect, social acts are on par with solitary mental acts such as perceiving or remembering, which are also irreducible to other mental acts. On a weak dispositional reading, Reid may be said to accept that all human beings are equipped with the relevant faculties that enable them to engage in social operations. In particular, it can be said that humans are always equipped with both the faculty to engage in a certain social act and its counterpart faculty, which enables them to react to the social act. For instance, this means that humans not only can make promises but also can accept them. A defender of the weak dispositional reading can argue that Reid's view does not require that the other intelligent being towards whom a social operation is directed always reacts to it by exercising the counterpart act, but rather it is sufficient that both intelligent beings are equipped with the relevant faculties, which enable them to engage in social operations and to react to them. However, does this weak reading accurately capture Reid's view?
Based on the passages cited earlier, I take it that Reid endorses a stronger view. As stated earlier, a promise must not only be expressed but also accepted by the other party. More generally, if intelligent being A initiates a social operation by making an act of expression directed at another intelligent being B, Reid holdsso I arguethat B has to react to it in each instance and make their reaction known to A to make the operation a social operation. Two clarifications are worth adding. First, Reid accepts that B can voluntarily decide how to react. Second, he further acknowledges that expression can take a variety of forms and can sometimes involve silence. Let me elaborate on each of these issues in turn.
On the interpretation that I am proposing, a social operation presupposes the existence of two (or more) intelligent beings who are equipped with the mental faculties that enable them to engage in the social act and the relevant counterpart act as part of their constitution. When one of the intelligent beings initiates the social operation and directs an act of expression to the other intelligent being, the other being is ready to exercise the counterpart operation but voluntarily considers how to react. Although in ideal circumstances, the reaction to a promise is the acceptance of the promise, acceptance of promises is not an automatic response. 23 Rather as an intelligent being, one has a voluntary choice whether or not to accept a promise. 24 Similarly, when one person commands something, the other is not compelled to obey the command, but rather is free to obey it or not. 25 Or, when someone asks a question, the addressee needs to use their intelligence to decide how to respond. This means that the exercise of one social operation prompts the exercise of the counterpart operation, but the content of the reaction is not determined. The important point is that the social operation, such as a promise, does not come into existence until the other intelligent being who takes part in it has exercised the counterpart mental act, which in this case is the mental act of accepting the promise, and made the response known to the person who initiated the social operation. 26 To turn to the second issue, although Reid holds that every social act must be expressed, because the act of expression is essential for communicating the mental thoughts between the two parties that take part in the social operation, it is worth noting that he acknowledges that expression can take a variety of forms. His lectures and papers on Practical Ethics include reflections on contracts that speak to this point: But it is to be observed that the Consent which is essential in all Contracts may be expressed many different ways: either by a formal writing Signed sealed and delivered; or by the verbal declaration of the several parties; or by the actions of the parties; or even sometimes by Silence, or by their doing nothing, when it may reasonably be presumed that they would not be silent or inactive if they did not consent. (Reid, 2007, XV, p. 136) The important point is that both parties understand the meaning of the signs that are used to express the content. Yet the signs do not have to be expressed by artificial written or spoken words and could also be natural signs such as bodily gestures, facial expressions, or modulations of voice. 27 To further illustrate the claim that silence can in certain circumstances be a form of expression, Reid gives the example of a parliamentary session, where members of parliament have the option to speak up if they disagree with a proposed motion. If they remain silent, their silence implies consent. However, he makes clear that silence in other circumstances can mean nothing. For instance, silently listening to a sermon at church does not reveal whether the hearer affirms what is preached. 28 Because the passage quoted here focuses on contracts, it is worth asking whether Reid's view that expression can take a variety of forms also extends to other social operations. He distinguishes promises from contracts as follows: In a promise, one party only comes under the obligation, the other acquires a right to the prestation promised. But we give the name of a contract to a transaction in which each party comes under an obligation to the other, and each reciprocally acquires a right to what is promised by the other. (EAP, V. 6, p. 328) This suggests that in a contract both parties make a promise and both parties have to accept the promise of the other party, while in a promise one party makes a promise and the other party has to decide whether or not to accept it. Because expression is not only an essential aspect of contracts, but also of promises, and more generally of all social operations, we can assume that Reid accepts further that the expression by words or signs can take a variety of forms with regard to all social operations.
The success of a social operation depends on the willingness of another intelligent being to take part in it. Because intelligent beings can voluntarily decide how they react to a proposed social operation, one may worry that there is a risk that social cooperation is diminished or undermined if intelligent beings refuse to participate too often. For instance, if promises are too rarely accepted, social interaction can fail; if reasonable commands are not obeyed, the functioning of society can be at risk; or if testimony is not believed, it can become difficult to acquire sufficient knowledge. Indeed, Reid acknowledges these concerns and states that 'if men found in experience, that there was no fidelity on the other part in making and in keeping [declarations and promises], no man of common understanding would trust to them, and so they would become useless' (EAP, V.6, p. 334). However, he is not too troubled by such worries, because he believes that God has made wise provisions to ensure that humans not only commonly speak the truth and are truthful to their promises, but also that they trust others to be truthful and to keep their promises. He already acknowledges this point in his early work An Inquiry into the Human Mind: The wise and beneficent Author of Nature, who intended that we should be social creatures, and that we should receive the greatest and most important part of our knowledge by the information of others, hath, for these purposes, implanted in our natures two principles that tally with each other.
The first of these principles is, a propensity to speak truth, and use the signs of language, so as to convey real sentiments. This principle has a powerful operation, even in the greatest liars; for where they lie once, they speak truth a hundred times. Truth is always uppermost, and is the natural issue of the mind. It requires no art or training, no inducement or temptation, but only that we yield to a natural impulse. (IHM, VI.24, p. 193) Another principle implanted in us by the Supreme Being, is a disposition to confide in the veracity of others, and to believe what they tell us. This is the counter-part to the former; and as that may be called the principle of veracity, we shall, for want of a more proper name, call this the principle of credulity. It is unlimited in children, until they meet with instances of deceit and falsehood: and it retains a very considerable strength through life. (IHM,VI.24,p. 194) He develops this view further in his Essays on the Active Powers of Man. There he identifies fidelity (principle of veracity in IHM) and trust (principle of credulity in IHM) as fundamental principles that make society possible: Hence it appears, thirdly, That this power of giving testimony, and of promising, can answer no end in society, unless there be a considerable degree, both of fidelity on the one part, and of trust on the other. These two must stand or fall together, and one of them cannot possibly subsist without the other.

PACIFIC PHILOSOPHICAL QUARTERLY
Fourthly, It may be observed, that fidelity in declarations and promises, and its counter-part, trust and reliance upon them, form a system of social intercourse, the most amiable, the most useful, that can be among men. Without fidelity and trust, there can be no human society. There never was a society, even of savages, nay even of robbers or pirates, in which there was not a great degree of veracity and of fidelity among themselves. Without it man would be the most dissocial animal that GOD has made. His state would be in reality what HOBBES conceived the state of nature to be, a state of war of every man against every man; nor could this war ever terminate in peace. (EAP, V.6, p. 334) From these observations, I think, it appears very evident, that as fidelity on one part, and trust on the other, are essential to that intercourse of men, which we call human society; so the Author of our nature has made wise provision for perpetuating them among men, in that degree that is necessary to human society, in all the different periods of human life, and in all the stages of human improvement and degeneracy. (EAP, V.6, p. 335) These two principles, namely, fidelity and trust, are an inherent part of our human constitution and guarantee that we are naturally inclined to speak the truth, be truthful, and to trust others. They can be said to provide a foundation for the various other social operations such as promising and accepting promises, testifying and testimonial belief, or commanding and obeying. Because fidelity and trust are fundamental principles of our constitution, they increase the willingness of intelligent beings to cooperate in social operations, provided that the proposed social operation does not conflict with rationality and moral conscience.
To sum up, I am proposing that Reid offers the following account of social operations: (i) The performance of social operations necessarily requires the existence of two (or more) intelligent beings who take part in the operation. Let us call them A and B. (ii) A social operation consists in A engaging in a mental act with a particular content directed at B. (iii) The mental act (e.g. the act of promising) does not come into existence until A has expressed the proposed social activity to B by words or signs (e.g. 'I promise you to φ') and B has understood it and expressed their reaction to A by words or signs (e.g. 'I accept the promise'). (iv) When B reacts to A, B also engages in a social operation, which is the counterpart operation to the operation that A initiated, and exercises a counterpart mental act (e.g. accepting promise), which is paired with the mental act exercised by A.

Further clarifications
It is worth reflecting further on the consequences of the proposed reading. If my reading is correct, it follows that a promise to oneself is not a promise, because promises are social operations and hence require the existence of another intelligent being who takes part in it. Because a self-promise does not involve another intelligent being, Reid has to accept that self-promises are not promises. Although it is not uncommon to regard promises to oneself as a special kind of promise, I take it that Reid would accept the consequence and argue that promising and self-promising are different mental acts. First, there are plausible philosophical reasons for distinguishing promises, which involve proper social interaction, from self-promises, because self-promises are not binding in the same way as genuine promises are. If promisor and promisee are identical, then the promisee can release the promisor from fulfilling the promise. 29 By contrast, if two distinct intelligent beings engage in a promise, then the promisee's acceptance of the promise makes it binding and confers a right on the promisee. 30 Second, there is textual evidence suggesting that Reid would argue that operations that we colloquially describe as a promise to oneself are better understood as a 'fixed purpose' or 'resolution'. 31 32 It is particular if the object of the resolution or fixed purpose is 'an individual action, limited to one time and place' (EAP, II.3, p. 66). By contrast, it is general if its object concerns 'a course or train of action, intended for some general end, or regulated by some general rule' (EAP, II.3, p. 66). Reid holds further that a resolution continues for as long as the subject wills it and 'he may at any time change his resolution' (EAP, II.4, p. 70). Consequently, it is plausible to assume that Reid would happily accept that self-promises are not promises and argue that they are better understood as resolutions or fixed purposes. By the same token, Reid would distinguish commands from selfcommands. With respect to commands, he holds that 'the object of a command is some action of another person, over whom we claim authority' (II.1, p. 49). He further notes 'that a command is a social act of the mind. It can have no existence but by a communication of thought to some intelligent being; and therefore implies a belief that there is such a being, and that we can communicate our thoughts to him' (EAP, II.1, p. 50). By contrast, when he speaks of 'self-command', he refers to a power of self-government, which humans have, but non-human animals lack. 33 I want to turn to another issue, namely, the question of what role the reaction of the other intelligent being plays in social operations. Put differently, the question is why condition (iii) is needed in addition to (ii) in the analysis given at the end of Section 2. As we have seen earlier, Reid holds that it is important that a promise is accepted by the other intelligent being. 34 Indeed, he claims that the act of acceptance is essential to 'make[… the operation] a promise' (EAP, V.6, p. 342). However, one may wonder whether condition (iii) extends to all other social operations. For instance, let us consider the social operation of asking a question. Does the operation of asking a question always require a response from the hearer of the question? Although one can utter a question in the absence of any intelligent being, for Reid, such an act of expression does not constitute a genuine question. To illustrate that at the very least it is essential that the other intelligent being understands the question, let us consider the following scenario. Assume that one intelligent being only speaks Italian and the other intelligent being only speaks Japanese. Let us assume further that the Italian speaker asks the Japanese speaker in Italian 'Che ore sono?' and the Japanese speaker does not react at all, because they do not understand the meaning of the utterance and do not recognize that the Italian speaker is trying to initiate communication with them. In this case, no social interaction has taken place. 35 Although there has been an act of expression, the social operation of asking a question does not come into existence.
Next, one may wonder why it is not enough that the other intelligent being hears the question and understands it. Questions can take a variety of forms. Some questions can be answered with a short 'yes' or 'no' answer, while others have one-word answers, and others again answers of one or several sentence(s). Some questions have one correct answer, while others can be answered in multiple different ways. Reid's view neither requires that the hearer can always give the correct answer, if the question has one correct answer, nor that the hearer gives a long response, if a full response would require a certain level of detail. For instance, if I ask a friend 'What is the capital of Switzerland?' and my friend responds 'I don't know', this will be sufficient for saying that the friend has reacted and participated in the social operation. Or, if I ask 'Why are some people happier than others?' and my friend responds 'I need to think more about it before I can give a proper response', this will constitute a response to the question even if it is not the response that the person asking the question was hoping to receive. It is worth noting that the response to a question does not have to be verbal. My friend could have also signaled through facial expressions or bodily gestures that she has understood the question, but does not know the correct answer or needs time to think more about it. The important point is that the intelligent being to whom the question is addressed signals willingness to engage with the question and this requires more than simply hearing and understanding the question.
I want to address another concern. One may find it surprising that A's mental act, say the act of promising, depends on other intelligent beings external to A's mind. However, rather than regarding this as a weakness of Reid's position, it is a consequence of his distinction between social and solitary acts of mind and his claim that social operations cannot be reduced to solitary ones. 36 There is clear textual evidence that for Reid social acts are distinct from other mental acts such as perceiving, hearing, remembering, judging, reasoning, or imagining and not reducible to them: To ask a question, is as simple an operation as to judge or to reason; yet it is neither judgement, nor reasoning, nor simple apprehension, nor is it any composition of these. Testimony is neither simple apprehension, nor judgement nor reasoning. The same may be said of a promise, or of a contract. (EIP, I. 8, p. 68) If the mental act of promising depended solely on A's mind, then it would be a solitary act of mind and Reid's distinction between social and solitary acts of mind would collapse and become redundant. Social mental acts still have a subject to which they belong. The mental act of promising is A's mental act, if A initiates the promise, and the mental act of accepting the promise is B's mental act, but each mental act is relational and depends on the participation of the other intelligent being.
To further clarify Reid's position, let us consider the following scenario: Assume that intelligent being A intends to make a promise directed at another intelligent being B. A may have already expressed the utterance 'I promise you to φ' in a letter addressed to B. However, unbeknownst to A the other person B has died in the meantime. Would A's utterance constitute a promise in such a case? Because Reid believes that a social operation necessarily requires that another person takes part in it, this is not a promise. In this case, an intention to engage in a promise and an act of expression exist, but these mental acts are different from the mental act of promising.
At this stage, it is helpful to distinguish the question of whether a promise has been constituted from the question of how A should act in the interim period until B accepts or rejects the promise or until A receives the news that B has passed away. In general, it will be prudent for A to act as if B has accepted the promise even if they do not yet know how B will react. Assume that A makes the following utterance directed to B: 'I promise you to keep this information secret'. If A does not receive an immediate response from B and shares the information while they are waiting for a response from B, then A will risk that the trust that others have placed in them is undermined. 37 Thus, it makes sense for A after they have uttered a promise to act as if B has accepted it. This is not because a promise has been constituted by the utteranceit is not constituted until B has accepted it and made this response known to Abut rather prudential considerations make it plausible to act this way. Moreover, as we have already seen, Reid believes that fidelity and trust are two fundamental principles of human nature and they can be said to explain why A most likely acts as if B has accepted the promise while A waits for a reaction from B.
A slightly different scenario concerns the possibility that the promisee dies after having accepted the promise. In this case, Reid would argue that the promisor is released from fulfilling the promise. The reason for this is that he acknowledges that each promise depends on a set of implicit external conditions, which are prerequisites for engaging in promises and being bound by them: No man can be certain of the continuance of any of his powers of body or mind for a moment; and, therefore, in every promise, there is a condition understood, to wit, if we live, if we retain that health of body and soundness of mind which is necessary to the performance, and if nothing happen, in the providence of GOD, which puts it out of our power. The rudest savages are taught by nature to admit these conditions in all promises, whether they be expressed or not; and no man is charged with breach of promise, when he fails through the failure of these conditions. (EAP, I.2, p. 18) This shows that there is no obligation to fulfill a promise if unforeseen external circumstances interfere with it.

Problems with Copenhaver's interpretation
In a very recent paper, Rebecca Copenhaver suggests that social operations are characterized by the following three necessary and jointly sufficient conditions 38 : (a) a belief that the operation is directed to a being that is intelligent; (b) the being to which the operation is directed is in fact an intelligent being; and (c) the operation be expressed to the being whom one believes to be intelligent, and who is an intelligent being. 39 I agree that the aforementioned conditions are necessary but question that they are jointly sufficient. 40 If we compare her analysis with the analysis I have given at the end of Section 2, one may notice that my analysis does not include her condition (a). This condition can be added without altering my interpretation. Indeed, I take it that it is already implicitly entailed by condition (iii). One problem with Copenhaver's analysis is that in the absence of further refinement it seems to include mental operations that can be directed towards another intelligent being, but are not necessarily so. To illustrate this point, consider two intelligent beings A and B and let us assume that A utters the following directed to B: 'I hope that you will win the lottery'. This utterance is accompanied by the mental act of hoping. The example satisfies Copenhaver's three conditions: A believes that B is intelligent, B is an intelligent being, and A expresses the operation, namely, their mental thought, to B. However, hoping is not necessarily directed towards another intelligent being and can be directed towards myself, say when I hope that I will win the lottery, or towards no intelligent being at all, for instance, when I hope that it will be sunny tomorrow. We have clear textual evidence that Reid would not classify hoping as a social operation, because it conflicts with his view that social operations 'necessarily imply social intercourse with some other intelligent being who bears a part in them' (EAP, V.6, p. 330). Instead hoping belongs to the solitary operations for Reid. This problem can probably be remedied. Indeed, Copenhaver cites the relevant passage where Reid states that by social operations he 'understand[s] such operations as necessarily suppose an intercourse with some other intelligent being' (EIP, I.8, p. 68), 41 but the modal operator 'necessarily' does not appear in her analysis (a)-(c). The point is that it is not sufficient that the operation is directed towards another intelligent being in a particular instance but that the operation necessarily requires another intelligent being who takes part in it. Although one could aim to revise the conditions Copenhaver has identified to better accommodate this issue, I will not do so here, because I believe that a deeper problem is that Copenhaver does not pay attention to the counterpart structure of social operations.
Let us focus on promises for a moment and assume that A and B are two intelligent beings, A believes that B is an intelligent being, and that A expresses 'I promise you to φ' directed to B. Given Copenhaver's view, I take it that she would accept that a promise exists in this case. However, this is not Reid's view, and the promise does not come into existence until the other person has accepted it. Let me illustrate this point, by turning to a scenario that involves temporal delay. Let us assume that at time t 1 A writes a letter to B and this letter includes the written expression 'I promise you to φ'. Let us assume further that B receives this letter a week later at time t 2 . A couple of days later at time t 3 , B writes a letter in response, which states 'I accept your promise'. A receives this letter a week later at time t 4 . This example makes it pressing to consider at what time the promise comes into existence. Copenhaver's interpretation is consistent with the view that it comes into existence at time t 1 . Although I am not aware of a passage where Reid directly engages with such a case, we have already seen that there is textual evidence that supports the view that a promise does not come into existence until the other intelligent being has understood it and accepted it. 42 For this reason, it is plausible to argue that the promise does not come into existence until t 3 or t 4 . Because Reid regards a promise as 'a social transaction between two parties' (EAP, V.6, p. 342), I take it that he would argue that the promise does not come into existence until time t 4 , because it is not until then that A knows that B has accepted the promise and knowing that B has accepted it makes the promise binding for A and confers a right on B. 43 Copenhaver does not acknowledge the counterpart structure of social operations. She focuses on the presence of two intelligent beings and on one party's expression of the mental operation but fails to acknowledge that the reaction of the other intelligent being is important additionally. This means that she neglects that the other intelligent being has to actively participate in the social operation to make it a social operation. My interpretation avoids the problems that arise for Copenhaver's view and is thus better suited to capture Reid's position.

Advancing Yaffe's interpretation
I now want to take a closer look at Gideon Yaffe's understanding of social operations. Yaffe ascribes the following view to Reid: Reid seems to be offering the following definition of a social act: a mental act is social if, and only if, the act's performance necessarily implies the existence of intelligent beings, other than the agent of the act, exercising their intelligence. (Yaffe, 2007, p. 282) Yaffe's definition is consistent with the account of social operations that I have given, but it is not as precise as the interpretation I proposed. In particular, Yaffe's statement that other intelligent beings are 'exercising their intelligence' is vague and lacks the level of precision that my view offers. Yaffe and I agree that social operations presuppose the existence of two (or more) intelligent beings who are exercising their intelligence. I am further aware that Yaffe regards the above as a definition and believes that it implies that a social act must be expressed. 44 This is because if A's performance of the mental act implies that another intelligent being B exercises their intelligence, then A must communicate with B through an act of expression. Hence, Yaffe's view can accommodate that A must express the planned social operation to B, but Yaffe's view does not include that it is further important that B expresses their reaction to A. Moreover, he does not make explicit that the mental act that B exercises when B exercises their intelligence is a counterpart to the mental act that A exercises. My interpretation has the advantage that it more clearly acknowledges the counterpart structure of social operations and thereby offers a more precise understanding of Reid's account of social operations. This is relevant, because Reid believes that we have distinct faculties, which are part of our human constitution and some of these faculties enable us to engage in social operations.

Reid's criticism of Hume and a possible Humean response
I want to end by assessing Reid's criticism of Hume's view that fidelity to promises is an artificial virtue and by considering a possible response on Hume's behalf. Before I turn to the evaluation of Reid's criticism, let me briefly introduce Hume's view. According to Hume, a virtue is a quality of mind that generates feelings of approval when contemplated from an unbiased point of view. 45 He maintains that there are artificial virtues in addition to natural virtues. Natural virtues such as loving one's children are part of our human constitution and neither education nor social conventions are needed to bring them about. However, Hume argues that we are not always naturally inclined to act virtuously and that there are virtues, which he calls 'artificial'; these virtues are human inventions and based on the needs and interests of society. He regards fidelity to promises as one of the artificial virtues. To illustrate Hume's thinking, suppose that you promise your partner not to drink in the evening. Hume asks what explains why you are willing to observe the promise. Assuming you are someone who enjoys a few drinks, you are not someone who is naturally inclined to refrain from drinking, but rather the reason why you are willing to make the promise may be that you have to drive home and you are aware that drunk driving is dangerous and could lead to severe penalties. Examples such as this prompt Hume to argue 'that promises are human inventions, founded on the necessities and interests of society' (Treatise, 3.2.5.7; SBN 519). 46 Reid's account of promises and more generally his account of social operations is meant to provide an alternative to Hume's position. Reid attacks Hume's view that fidelity to promises is an artificial virtue by accusing Hume of failing to realize that there are social operations in addition to solitary ones. Reid believes that by acknowledging that promising is a social operation he can account for its naturalness. Let us now consider whether his view offers a satisfying response to Hume.
Reid's account of social operations is sophisticated; it draws on his analysis of the human mind and his thesis that it is important not only to distinguish different contents or objects of mental operations but also to distinguish the different constituents of mental operations, which include mental acts in addition to the contents or objects. Reid's theory of the human mind includes a multiplicity of different mental faculties, which produce distinct and simple mental acts such as perceiving, remembering, imagining, reasoning, testifying, and promising, which are distinct from each other and not reducible to mental acts produced by other faculties. Realizing that social operations involve irreducible mental acts and that the relevant mental faculty is an inherent part of human nature is at the core of Reid's view for why social operations are natural. This leads to the question of whether Hume would be willing to accept Reid's move.
Hume has limited resources to account for distinct mental faculties. He appeals to different degrees of force and vivacity to explain the differences between sense perception, memory, and imagination. Hume claims that the contents of perceptionwhich he calls impressionsare present in the mind with full force or vivacity, while the contents of memorynamely, ideas in Hume's senseare less lively, and when ideas entirely lose their force and vivacity, they are contents of the imagination. 47 Reid finds this view unconvincing and challenges it with the following example. He claims that a human head can touch a wall with different degrees of force: First, a head can bump into a wall with strong force; second, it can strike the wall with medium force; and, third, it can touch the wall gently. Reid's point is that it does not make sense to regard the first way of touching the wall as perception, the second as memory, and the third as imagination, as Hume would have it. 48 Moreover, Reid argues that Hume's philosophical system lacks the resources to properly differentiate between different types of mental acts. 49 Does the fact that Hume cannot easily accommodate Reid's distinctions between different mental faculties or types of mental acts mean that Reid has already won the argument? Although Reid's distinctions certainly help overcome ambiguities in Hume's (and Locke's) philosophical works, it would be too quick to dismiss Hume's position solely on this ground. To better assess whether Reid's criticism of Hume's account of promises is convincing and fair, I will now outline a possible response that Hume could offer.
When Hume argues that fidelity to promises is an artificial virtue, he focuses on the content of promises and is interested in understanding what motivates a promisor to observe a promise with a certain content. This question is particularly pressing in cases where the promisor has conflicting desires. Reid, however, does not tackle this issue head on and instead shifts the focus of the debate towards the mental act of promising and its accompanying act of expression and argues that the mental faculty to engage in acts of promises is an inherent part of human nature and, thus, natural rather than artificial. Because Reid believes that intelligent beings can voluntarily decide whether or not to engage in social operations and can voluntarily decide how they exercise the counterpart operations, Reid owes Hume a better response to his question. For instance, consider an example where one intelligent being A faces a choice between two possible promises. One option is to promise their mother B to eat cake; the other option is to promise her not to eat cake. In cases such as these, context can be relevant. For instance, it is possible that A would much prefer to do other things than join the family to eat cake. Or another possibility is that A is on a diet and for this reason should not eat cake. In either scenario, the question is pressing why A is willing to make and keep the promise in question, because A could have also promised the opposite or made no promise at all. Reid argues that it is natural for humans to engage in the act of promising, but he has not said enough to explain what, for instance, motivates one to promise to φrather than to promise not to φ or not to promise something at alland to keep this promise.
Reid not only claims that our faculty to engage in promises is natural, but he further argues that God has equipped all humans with two fundamental principles, namely, fidelity and trust, which are inherent parts of our human constitution. Assuming that these principles are constituent parts of human nature, Reid argues that humans are naturally inclined to keep their promises. Yet this response also remains silent on Hume's concern regarding the content of promises. Especially, why would someone make and keep a promise if its content conflicts with one's desires? Although Reid does not engage with this concern in depth, he may be less worried about it than Hume, because Reid believes that as moral agents humans will be guided by rational principles and moral conscience. If the content of a promise conflicts with rational principles or moral conscience, then there is good reason not to make it or not to accept it. Overall, Reid just is more optimistic than Hume about the willingness of human beings to engage in social operations. For Reid, this involves that humans remain truthful to promises and trust others. 50 School of Philosophy & UCD Centre for Ethics in Public Life University College Dublin NOTES by chapter, section, and page numbers. For Reid, signs and the things they signify have to stand in 'a real connection' (IHM, VI.21, p. 177) to each other, but the relation between them is not one of metaphysical necessary connection nor does it require resemblance between the sign and the thing signified and is best interpreted as regular correlation. Artificial signs are conventional and 'have no meaning, but what is affixed to them by compact or agreement among those who use them' (IHM, IV.2, p. 51). For instance, the English word 'lemon' is an artificial sign that signifies lemons, but it had no meaning until it was agreed that this expression is to be used to refer to lemons. By contrast, natural signs do not require conventions or agreements and 'have a meaning which everyone understands by the principles of his nature' (IHM, IV.2, p. 51). Reid distinguishes three classes of natural signs. The first class 'comprehends those whose connection with the thing signified is established by nature, but discovered only by experience' (IHM, V.3, p. 59). Examples of such natural signs include smoke as a sign of fire (see IHM, VI.21, p. 177). The second class concerns 'that wherein the connection between the sign and the thing signified, is not only established by nature, but discovered to us by a natural principle, without reasoning or experience' (IHM, V.3, p. 60). This class includes bodily gestures, modulation of voice, and facial expressions such as a smile as a sign for happiness. The 'third class of natural signs comprehends those which, though we never before had any notion or conception of the things signified, do suggest it, or conjure it up, as it were, by a natural kind of magic, and at once give us a conception, and create a belief in it' (IHM, V.3, p. 60). Reid is here thinking about examples such as sensations as a sign of 'a sentient being or mind to which they belong'