社會契約概論
「功利主義」(Utilitarianism)是西方倫理學的中心思想。即使多年來面對眾多批評及挑戰，其影響力仍然歷久不衰。另一方面，自羅爾斯(John Rawls)在1971年出版了《正義論》(A Theory of Justice)一書後，沉寂已久的「社會契約論」(Social Contract Theories)（下稱「契約論」）又再次為人所談論。作為一種非功利主義形式的理論，「契約論」代表了一個與所有後果主義理論(Consequentialism)相對的體系。可惜的是，這龐大而且複雜的體系包含了一系列「貌合神離」的理論，它們對什麼是「契約」、契約有什麼作用等問題都有各自的理解，使整個體系表面上缺乏了統一性和完整性。本文的目的，正在於釐清「契約論」的一些基本概念，並對不同類型的契約理論作一概括的分析。文章的第一部份會略述「契約」的意義；第二部份會提出一個理論的框架，我將以之分析「契約論」的基本結構。
（一） 何謂「契約」？
「契約論」作為一規範倫理學的學說，目的不僅在於提出一套基本的規範性原則，更在於為這一套原則作出證立。不管這套原則的內容是什麼，所有契約理論都以「同意」、「承諾」等與「契約」有關的概念作為其理論的基礎。因此，不同契約理論之間的共同之處不在於它們要「證立什麼？」，而是在於它們「怎樣證立？」。雖然在細節上或有不同，但所有契約論者均認為，一套規範性原則之所以成立，其最終的證立必須訴諸「社會契約」。什麼是「社會契約」呢？這是一個社會上所有人共同參與的契約，不但為所有人所共同制定，更為所有人所承諾遵守，是故約束力及至社會上每一位成員。[1]正是由於能為每一個社會成員所接受，那一套原則才能得到證立，它的規範性由此而來。理論上這是個非常吸引的論證：假設一套原則在某些意義下為全部社會成員所同意，其權威性好比一個得到全民投票支持的政府。有一點要特別留意的是，契約論者和功利主義者可以同意哪一套規範性原則最終能得到證立，例如大家都可以同意道德的基本原則在於「公平對待所有人」，但雙方會採用不同的證立方法，前者會認為這原則成立的理由在於它為「社會契約」所確立，後者則認為它符合「功利原則」──因為遵守它能帶來較大的社會效益，所以它才成立。這是這兩個理論最基本的不同之處。。
為什麼契約對契約者有約束力呢？由於這裡所說的契約並非一法律概念上的契約，因此契約對契約者具有約束力並不是因為他們有法律上的責任去遵守它。要回答這個問題，我們必須找出一些道德上契約者必須遵守契約的理由。一般來說，契約之所以具有約束力，原因在於它代表了契約者的一項「承諾」(Promise)，就如答應了別人要做一件事，我們便有責任實踐這個諾言。「承諾」本身也蘊涵了契約者的「同意」(agreement): 因為大家也同意受契約所約束，所以契約才行之有效。對契約而言，「承諾」的行為較為明顯(explicit)，無論是口頭上的答應或白紙黑字的約定，它的表達形式也較為清晰；相反，「同意」則較為隱晦(implicit)。
（二） 「契約論」的基本結構
由於「契約論」認為某些具規範性的原則之所以得到確立，是因為某些人在某些特定的環境裡同意遵守它們，所以「契約論」的結構分析可以歸納為三個基本的問題：（１）契約的評價對象 (evaluative subject)問題，確立了的原則在在什麼範疇上適用？契約除了可約束我們的行為之外，可不可以應用於政治、經濟、法律及其他社會制度之上？（２）契約者 (contractors)的問題，契約由誰來同意？在誰的身上適用？契約者有什麼特徵？（３）契約環境 (contracting situation)的問題，契約在什麼初始條件的限制下達成？契約是實在的(actual)還是假設的(hypothetical)？透過這個分析的框架，我們可以看到不同的契約理論怎樣利用不同的方法建立其理論，它們之間的差異就在於它們為這三個基本問題所提出的不同答案。[2]
首先討論的問題是契約的評價對象，也就是上面所說「證立什麼？」的問題。如上所述，契約理論的主要功用是為一套規範性原則提出證立。社會上除了道德系統有其一套規範性原則之外，政治制度、法律制度、經濟制度等也可以有其自己的一套規範性原則。要理解一個契約理論的具體內容，首先便要知道這一個理論要證立哪一個範疇的規範性原則，簡單來說，就是要知道這個理論以什麼為評價的對象。比較經得起考驗的契約理論都只會以一個比較清晰的對象為評價中心，而不會過份廣泛地解釋所有的規範性原則。17－18世紀時期的契約論者，如霍布士(Thomas Hobbes)、洛克(John Locke)、盧梭 (Jean-Jacques Rousseau)等，大部份都將其契約理論應用在政府合法性的討論之上。例如霍布士的契約論，其目的主要是要證明在自然狀態的前題下，社會上的人會選擇放棄部份的自由而受政府所管轄，以此來證立政府的合理性，契約應用的對象因此只限於政府。[3]當時只有康德(Immanuel Kant)的契約理論比較直接地應用「契約」到道德原則的證立之上。
羅爾斯以契約理論建立的「正義論」，其評價對象是社會上的「基本結構」(basic structure)，包括最主要的政治體制及經濟制度。[4] 羅爾斯之後，隨著「契約論」的復興，開始有更多的契約論者以其他的評價對象作為其理論的中心。例如哈貝瑪斯(Jürgen Habermas)、葛地亞(David Gauthier)、史基倫(Thomas Scanlon)等均以契約理論來建立道德論證的基礎，他們的討論都主要集中在道德的範疇之內。但即使限制在道德的範疇之內，不同的契約理論還是可以有不同的評價對象。例如以「道德規條」為直接評價對象的契約理論，會認為一些道德規條如「不可隨便殺人」等之所以具規範性、有約束力，其最終的理由在於它們為社會上的契約者所同意遵守。在此一意義下，一個行為道德上是否可以接受，取決於它是否被這些為契約所確立的規條所允許。當然，其他契約理論也可直接以「行為」為評價對象，認為一個行為之所以在道德上可以接受，是因為它本身就已經為所有契約者所允許。在此意義下，一道德規條能否得到證立在於其是否容許那些為契約所確立的行為。相對於「規條功利主義」及「行為功利主義」，這兩種以不同對象為評價中心的契約理論可以分別稱為「規條契約論」及「行為契約論」。除了以「規條」或「行為」作為評價對象外，一個契約理論甚至可以以「品格」或「行為動機」作為其評價對象，例如認為一套品格是好的或一個行為動機是好的，其原因在於所有契約者均同意它們是好的。間接地，一個行為的好壞便可從其是否為這些契約所確立的品格或動機所衍生來判斷。理論上，我們甚至可以有更多的評價對象。雖然實際上未必真的有以這些評價對象為中心的契約理論，但「契約論」本質上是不會排斥任何事物作為其評價對象的。從以上的討論可見，「契約論」本身可以應用於很多不同的評價對象之上。要了解一個契約理論的具體內容，首先要理解其評價的對象。
「契約論」結構上的第二個基本部份是契約者。契約本身由什麼人所制定，將直接影響契約的內容，而不同的契約理論對契約者也會有不同的描繪。表面上，這個問題好像沒有什麼意義：假如要證立的是一個政治制度，契約者當然應該包括所有受這個制度約束的人。問題是，這些契約者擁有什麼特質？是什麼性格上、心理上、甚至生理上的特質使他們同意如此這般而不是別的契約？這可以稱為「契約的主觀條件」。假如所有受契約所約束的都是現實世界的人，這是否意味所有契約者都必須被理解成現實世界裡的人，擁有一般人的智力、心理及生理的特質？
某些契約理論的確採用這樣的進路，以真實的人作為理論中的契約者，將他們描繪成擁有一般人的正常特徵。如此一來，「契約的主觀條件」包括真實世界裡的人的智力、知識、價值觀及行為動機等，契約的具體內容將為這些特徵所決定。以「現實化的契約者」的選擇為基礎的契約理論，其理論的優點在於簡單直接。既然契約是由「現實化的契約者」所制定，我們只須想想在「正常狀態」下我們會同意什麼便可。例如哈貝瑪斯的「語境倫理學」就是以完全現實化的人作為契約者，以他們在理想的客觀環境下作的選擇為理論的基礎。但這一類型的契約理論同時也有一個很大的困難：何謂人類的「正常狀態」？究竟現實上的人擁有什麼特質？根據經驗上的觀察，有些人天生便有超乎常人的智力，有些人則連正常的智能也沒有；有些人自私自利，也有些人喜歡幫助別人；而更重要的是，不同文化背景下的人對何謂「好」、何謂「對」也有不同的理解。既然每個人在心智上、性格上、對世界的認知上、價值觀上及行為動機上都有所不同，我們怎樣能夠肯定他們在某些評價範圍內取得共識？即使有，假如我們不能清楚指出「現實化的契約者」的主觀條件是什麼，我們又怎能證立他們因此而選擇的那些契約原則具有規範性呢？因此之故，大部份的契約理論都會對契約者作出一定的假設。與「現實化的契約者」相對，這些契約者可統稱為「理想化的契約者」。
「理想化的契約者」有什麼特徵呢？不同的契約理論對他們都有不同的假設，大家的分別只在於理想化的程度不同而已。在智力上及知識上，有些理論假設契約者是「智力正常」的，同時擁有一般人的常識，例如知道某一個決定通常會產生什麼後果、對自身有什麼影響等；有些甚至假定他們擁有超乎常人的智慧，對現實世界的因果關係瞭如指掌；有些理論則從反方向假設契約者對某些事情無知，例如假設他們對自己的身份及社會地位並不知情，從而保證他們在立約的過程中不會堅持某一個立場。在價值觀上及立約的動機上，有些理論假設所有的契約者都是自私的，或都只追求自身利益的最大化；有些則假設他們在追求自利的同時，也會有關心他人的動機，甚至將他們的動機直接理解成追求普遍利益的最大化；另外一些契約理論則不從利益的滿足著眼，而將契約者的動機表達成一種對其他道德主體的關懷，或一種對和平共處、互相尊重的社會的追求。總而言之，不同的契約理論可按其對契約者的理想化程度來分類，「現實化契約者」理論及比較溫和的「理想化契約者」理論對契約的主觀條件較少假設，例如只假設部份(或全部)契約者都擁有正常智力，大部份時間理性而且擁有一般的知識，價值觀並不相同而立約動機多樣。比較極端的「理想化契約者」理論則對契約的主觀條件有較多的假設，例如可以假定契約者絕對地理性，對立約的條件、別人的動機、契約的影響等事實有全盤的認識，卻同時不知道自己的身份，並且只有一個或數個清楚的立約動機。不用多解釋就可看到，不同的契約理論之所以有不同的契約內容，部份原因就是它們對契約者及契約的主觀條件有不同的理解。
契約者的理想化帶出了另一個相關的問題。可否將契約者理解成完全相同的人呢？契約者在高度理想化之後，有機會變成在某些方面完全一樣的人，例如擁有相同的智慧、相同的立約動機甚至相同的價值觀。比較極端的理論甚至可以將契約者理想化為一個或多個擁有相同特徵的「理想觀察者」(Ideal Observer)，我們可稱這類型的契約理論為「相同契約者」理論。如此一來，契約者為什麼會同意契約的條文、契約的主觀條件怎樣決定契約的具體內容等也就不言而喻。當然，「理想化契約者」理論不一定要將所有的契約者都理解為同一樣的人，它們可以假定契約者在智力及知識上平等，但卻各自懷有不同的立約動機；也可以假定他們在動機上相同，但對現實的認識卻有不同，或處於不同的社會地位等等。由此可見，雖然「現實化契約者」理論基本上很有可能是「不同契約者」理論，但「理想化契約者」理論卻不一定是「相同契約者」理論。有人認為「不同契約者」理論仍然解決不了為什麼不同的契約者會達至相同的結論這個理論困難，這明顯是基於對「契約論」的片面理解，忽略了契約的主觀條件與客觀條件之間的互動。即使契約者在某些與立約有關的條件上並不相同，在有利的客觀契約環境之下，他們也有可能達成一致的結論。由此可見，在現實化與理想化、相同與不同之間，再配合上不同的契約環境，「契約論」可以有很多出人意料的變化。
最後一點關於契約者的問題是，契約的參與者是否必須是個人？雖然在西方個人主義的傳統背景之下，大部份契約理論裡的契約都是個人與個人之間的契約，以個人為參與者的單位，但本質上「契約論」其實並不排斥以其他單位作為契約者的可能性。例如在建構國際社會的正義原則時，羅爾斯就引用了社會與社會之間的契約的概念。[5] 我們也可以以其他群體作為契約的單位，建構例如國家之間、民族之間或社會階層之間的契約。我們甚至不必將契約者限制在同一時代的群體，例如在考慮環保政策時，我們可設想我們的下一代會與我們訂立什麼契約，他們會不會同意我們毫無節制地運用地球的資源？這是一代與一代之間的契約。「群體契約論」與「個人契約論」的分別在於，「群體契約論」不用訴諸參與契約的每一個人的同意或承諾，其理論基礎在於代表著個人的群體之間的協議。當然，群體的意願未必能充分代表個人的意願，而假若一個契約並不能得到所有契約者的同意或支持，它在理論上的優點將大打折扣。由於這是個別群體契約理論有待解決的問題，我將不會在此詳細討論。
契約論的第三個基本特徵是契約的環境。如果對契約者的描繪代表了「契約的主觀條件」，那麼對契約環境的理解則代表了「契約的客觀條件」。如上文所述，即使契約者因為在智能上、性格上、價值觀上的不同而沒有一致的主觀選擇，在某些特定的客觀條件下，他們也能達成大家也同意的協議。例如葛地亞的正義契約理論，將契約的客觀環境描繪成「囚犯兩難」式的境況，以此說明以滿足個人利益為目的的人，即使主觀意願不盡相同，在合作對大家都有利的情況下，都能在磋商的過程中達成協議。在此一意義下，契約的達成並不是因為契約者的主觀意願一致，[6]而是因為在客觀環境的規限下，所有契約者都會為達成協議作出讓步(Compromise)，從而得出某些共識。這亦解釋了為什麼某些理論會稱契約者為「交易者」(Bargainers)。
契約環境除了包含「契約的客觀條件」之外，廣義上還包括契約的性質。從契約的性質而言，契約可分為「真實契約」及「假想契約」兩種。以「真實契約」為基礎的契約理論，以契約者有否實際上同意或承諾某一套規範性原則來決定這套原則是否得到證立，例如選舉制度中有「票高者得」的原則，這原則之所以對所有參與選舉者都具有規範性，原因在於參與者實際上在參與選舉之時都已經同意了這項原則，不能因為選舉最後的結果不利自己而輸打贏要。相反，以「假想契約」為基礎的契約理論，不必訴諸契約者實際上同意了什麼原則，而是訴諸契約者在某些特定的假設環境之下將會達成的共識、或同意的某一些規範性原則。這項「契約」實際上並沒有達成，其理論的基礎並不在於契約者實際上同意了什麼，卻在於契約者在理想的客觀環境之下會同意些什麼。由此可見，「真實契約」與「假想契約」對規範原則的證立採用了不同的策略：前者較著重「承諾」的實踐，因為實際上同意了而有責任實踐承諾；而後者則以理想的「選擇」為重點，規範原則之所以成立是因為所有契約者在理想的環境下都會有這樣的選擇。
有一點需要特別留意的是：「真實契約」與「假想契約」的分別不能與「真實契約者」與「理想契約者」的分別混淆。前者是對契約環境的描繪，而後者則是對契約者的描述，因此即使很難想像「理想契約者」能達成任何「真實契約」，「假想契約」卻不一定由「理想契約者」所達成。我們可以為「真實契約者」設立假想的契約，設想這些沒有經過理想化的契約者在某些假設的客觀環境下會達成什麼協議。事實上，很多較為流行的契約理論都採用了「真實契約者」[7]與「假設契約」的組合，例如史基倫的倫理契約論，當中的契約者與常人分別不大，有著不同的價值觀、行為動機及智力，只假設了他們都有追求共同接受的原則(mutually-acceptable principles)的動機，而倫理原則的確立就是建立在這些接近真實的契約者在假想的契約下所達成的共識之上。[8]
。
這裡可籍羅爾斯的正義理論來展示以上的分析架構。如上所述，正義兩原則所評價的對象是社會的基本結構，例如經濟制度怎樣分配資源與利益、政治制度怎樣分配權利與義務等等。在無知之幕(veil of ignorance)的假設之下，契約者被描繪成對自己的身份、社會地位、長處與短處、甚至是價值觀一無所知，而推動他們去達成契約的動機則是互不關心的理性(mutually disinterested rationality)。眾所周知，證立正義兩原則的契約並非事實上的契約而是假想契約，它的重點在證明兩原則是契約者在無知之幕下理性的選擇，因此其重點在於「同意」而不在「承諾」。關於羅爾斯正義論的學術文獻甚多，在此不贅。總括而言，羅爾斯之後重要的契約論，例如史基倫的contractualism及的葛地亞的contractarianism等，都可以以上的架構分析。
[5]關於羅爾斯的國際正義理論，可參考Rawls, The Law of Peoples, The idea of public reason revisited (Cambridge, Mass., HarvardUniversity Press, 1999)。
[8]史基倫的契約環境假設了每一個契約者的考慮都會得到獨立、公平的計量(separate and fair consideration for each individual’s claim)。關於他的契約理論，可參考T. M. Scanlon, What We Owe to Each Other (Cambridge, Mass: Harvard University Press, 1998) Ch. 4。
Contract Theories: Critics
Critics for contract theories in general are presented in the form of a dilemma: the contracts employed in contract theories are either actual or hypothetical, but both types of contract have its fatal weakness and therefore contract theories as a whole is not as plausible as it seems to. For actual contract theories, obviously, the main problem lies in the fact that we have never engaged in any agreement historically or actually, either in justifying our states, our moral systems or our social institutions. Even if the contract is agreed upon actually by our ancestors in the past, it is not sure why people today, who were not parties of that original contract, would be bound by arrangements to which they personally have never agreed. Moreover, actual contracts agreed in actual circumstances always involve morally irrelevant and morally undesirable features. Why is a contract binding if it is not agreed deliberately, rationally and voluntarily? Everyone would agree that we are not obligated to fulfill a promise when it is made at gunpoint or when we are deceived of certain essential information in making the contract. The mere fact that a contract is agreed upon does not necessarily mean that it contains morally binding terms.[1]
These shortcomings move contract theories to two different directions. To answer the charge that actual contracts are factually unfounded, we can shift to an implicit version of contract theories. To answer the charge that actual contracts may contain morally undesirable features, we can move to adopt a hypothetical contract. I will start with the argument for an implicit version of contract theories. While rights and responsibilities are clearly defined and the acceptance of which are positively indicated (usually in terms of signing it) in explicit contracts, details of the contract terms are not fully spelled out in implicit contracts. Implicit contracts do not possess the open, concise and definite character of an explicit contract. It is argued that although no contract is made explicitly in actual circumstances to justify our states, moral systems or social institutions, we have implicitly agreed to these practices by continued participation in, and acceptance of benefits from, these practices. If an agent knowingly receives benefits from a mutually beneficial scheme and continues to participate in it, it signifies the agent has implicitly engaged in an agreement to that scheme. It is obviously ridiculous to claim that a diner can refuse to pay the restaurant after having a meal because he has not 'actually' agreed to pay. Similarly, states, moral systems or social institutions could be justified if they are implicitly agreed by us in terms of a social contract.
Implicit contracts are still actual contracts although they are not explicitly acknowledged. They are also actually agreed by us, just having a more liberal conception of what can be counted as a valid agreement. For example, continued participation in a contract by abiding to its terms or knowingly and freely receiving benefits from a contract is claimed to signify an actual agreement to that contract. Sometimes, it is even argued that all explicit contract theories should best be understood as implicit contract theories. The interpretation of any agreement, even an explicit one, depends on conventions of interpretation, and it is simply impossible to make all the terms of a contract explicit. Moral debates can be understood as further implicit moral bargaining or as making explicit that which was already implicit in a moral agreement.[2]
At least two problems can be identified for implicit contract theories. First, the basic assumption of implicit contract theories is that if contractorsof the agreements are unsatisfied with an implicit contract, they can attempt to break it or re-negotiate it simply by discontinuing to participating in it or by refusing to accept its benefits. If they actually disagree to the contract, they could just stop engaging in the contract any more. This assumption is doubtful. Does it mean whenever I disagree with the policies of the state, I have to leave the country to signify my disagreement? Does it mean whenever there is an unjust law, civil disobedience is a must? Not explicitly disagree does not mean implicitly agree. Continued participation and acceptance of benefits is just an arbitrary standard for implicit agreement. The basic problem of implicit contract theories is that it is not clear what behaviors indicate an implicit consent to particular arrangements. If we have a low threshold for 'implicitly agreeing to a certain contract', for example, as it is argued by John Locke that using public roads indicates an implicit consent to the government,[3] then it is doubtful whether all the contractors actually know that they have engaged in a contract when they are doing so. It is also doubtful whether this kind of 'unconscious' agreement does have strong moral force on us. If we have a high threshold for implicit agreement, perhaps by requiring the contractors to have a collective agreement expressively in a public forum, then it is not sure whether there can be any determined agreement at all. They may still suffer the charge that they are factually unfounded. When there is no consensus on what behaviors indicate an implicit agreement, implicit contract theories are not initially convincing.
Second, even if it is accepted that continued participation and receiving of benefits indicates an implicit agreement to a particular contract, it cannot avoid another problem of actual contract theories: that it involves morally undesirable and morally irrelevant features in making the contract. Take political contract theories as an example. Sometimes people may implicitly 'agree' to a political arrangement just because they cannot depart the country due to actual constraints (perhaps they are too poor or too weak to do so), or due to their ignorance of any better alternative political arrangements. In the former case, the contractors are forced to accept the contract just because they have no other better choices while in the later, they are deprived of certain important information about the contract. Therefore although they have 'agreed' to the contract, perhaps implicitly by continued participation in and acceptance of benefits from the arrangement, the binding power of the contract could be challenged. It should be emphasized that implicit contracts, just like actual contracts, involve no idealization of the contracting situation. If it is the case, they cannot avoid the undesirable or irrelevant features in actual contracting situations.
If implicit contract theories are not satisfactory, we can move to a hypothetical version of contract theory. Hypothetical contract theories are particularly designed to deal with the problem concerning morally undesirable and morally irrelevant contracting situations of actual contracts. A hypothetical contract theory can remove these morally undesirable and irrelevant features by stipulating a more suitable contracting situation. It is even argued that justification can only come from what is agreed upon in an idealized contracting situation, but not what is agreed in an actual situation which may be unfair or compulsory. In this sense, hypothetical contracts are not just proxies for actual contracts; they bypass and override actual contracts.
However, it is not cost-free to employ an idealized contracting situation. Critics claim that idealization is illegitimate and arbitrary, and they are just begging the question of what is relevant in contracting situations. Certain features must be presupposed to define a 'suitable' contracting situation. If not, we simply do not know when a binding contract is made.[4] For example, a theory can define the suitable contracting situation as one that is fair and equal. 'Fairness' and 'equality' are all moral notions and it is claimed that as a foundational theory of morality, contract theories should not employ any moral notions as implicit assumption. These assumptions are not beyond question and it is illegitimate to take them for granted at the first place.
The force of this criticism depends on whether the contract theory employing these moral notions tries to explain all moral notions within a unified framework of a single contract. So far, if the theory does not claim to serve as the foundations for all moral notions, this criticism would miss the point. Whether a theory is in fact guilty of circularity depends on whether there is any further, adequate and independent argument for those assumptions. A theory aiming at identifying the basis of a certain normative factor, say 'justice', is not circular nor illegitimate if it draws 'justice' from other normative notions like 'fairness' or 'equality' if we can justify these notions independently. The most important point is whether we have a shared ideal for deliberation about the contracting situation for that particular normative factor. Here, we can appeal to our considered convictions in reflective equilibrium in justifying what are the legitimate bargaining situations for different normative factors of different evaluative focal points. A hypothetical contract theory cannot be dismissed simply because it employs some assumptions that it cannot account for.
Even though it is not illegitimate to have presupposed moral assumptions, appeals to them may still invite the charge that hypothetical contracts are unnecessary or trivial. This criticism usually comes in the form of the Euthyphro dilemma: is something legitimate because it is agreed upon in the contracting situation, or is it agreed upon in the contracting situation because it is legitimate?[5] On one hand, if deliberations in the contracting situation are not guided by substantive moral standards, then we may not be able to get any determinate agreement from the contract; and deliberations in the contract, if any, would be arbitrary because they are not supported by reasons. On the other hand, if deliberations in the contracting situation are guided by substantive moral standards, then it is these standards (not the fact that the bargainers would agree to the contract) that give the deliberations moral force. The contract is agreed upon because it is legitimate, not the other way round. Agreement in a contract can just serve as a heuristic device in identifying what is legitimate but cannot provide the foundation for it. Appeal to what is agreed upon is therefore unnecessary because what is legitimate could be justified directly by appealing to those substantive standards that guide the contractors' choices. If this is true, contracts are just ad hoc devices; they cannot derive any moral force from themselves.
Two responses are available for hypothetical contract theories. First, the Euthyphro problem is a problem faced by all foundational theories but not solely by contract theories, and it does not pose an attack to those heuristic versions of contract theories. Second, whether agreements in contracts are trivial depends on whether the choices are determined directly by those substantive standards. Contract theories can argue that the idea of agreement is still basic in deriving moral force of the contract: what is legitimate is still because it would be agreed, and those standards just explain why the contracting situation could be agreed. So far, contract theories do not claim that what is legitimate is simply constituted by those substantive standards. They just claim that what is legitimate is determined by a contracting situation that employs these standards. We still have to appeal to what is agreed to in determining what is legitimate. For example, Rawls does not claim that justice is fairness but justice is derived from a fair procedure; and Scanlon holds that what is right or wrong is to be judged by principles derived from a reasonable situation, but not right or wrong is simply reasonable and unreasonable.[6] Contracts are not trivial in this sense because they represent a legitimate stage where justification can be derived. Those substantive standards explain why the contract is a legitimate process for deriving normative principles but cannot explain in itself why the justification derived from the contract is legitimate. It is the fact that the contract is agreed upon that makes it legitimate. It may be argued that these standards are unwarranted; but as I have argued above, it is unwarranted only when there is no independent argument for these standards.
Hypothetical theories can do a better work than actual theories in stipulating a more suitable contracting situation. However, it is doubtful whether it can avoid another shortcoming of actual contract theories: that they are factually unfounded. Hypothetical contracts are not actual contracts agreed by us as a matter of fact. It is normal to ask why, if hypothetical contracts are the ones that we have never actually entered into, we would have any obligations towards the contract. Ronald Dworkin has famously argued that you cannot reasonably argue that I should sell a valuable paintings to you for $100 on Wednesday, just because I would have agreed to sell it to you for $100 on Monday when I have not discovered the values of the paintings yet.[7] It is claimed that we are only bound by what we have actually agreed to.
The challenge to the binding power of hypothetical contracts is fortified if hypothetical contracts are combined with not so realistically characterized contractors. If a contract is agreed by hypothetical beings so idealized and different from us, why we have any reason to respect the principles that they, the idealized persons in a hypothetical condition, would adopt? These idealized contractors may be in a better position to deliberate, but they cannot represent the real choices of actual beings like us. Their radical differences with us may render their choices irrelevant to us.
The charge that hypothetical contracts are not actually made assumes as if the real force of hypothetical contract theories derive from the importance of keeping promises. They just treat hypothetical contracts as poor substitutes for actual contracts when there is no alternative, just like appealing to what a dead person would agree to in distributing his assets when he had not specified his will before. However, as I have argued above, hypothetical contracts are not just second-best proxies for actual contracts. They represent genuinely better perspectives in deliberating legitimate moral principles. If we agree that legitimate principles should be derived from a legitimate position, then it is odd to suggest that such position should be one that is unfair, irrational, or full of prejudices, just like the one we face in actual situations. The real force of argument from a hypothetical contract does not lie on the fact that we have agreed, but on the fact that we would have agreed with good reasons. The principles derive from a legitimate contract represents what the contractors actually have reasons to accept, not only what they would have reason to accept if they were so and so. The fact that we would have agreed to do something under idealized situations therefore provides an independent argument and a prima facie reason for us to be bound by it.[8]
The problem of irrelevance of idealized beings to us can also be reduced if contractors are characterized in a fairly realistic way. If contractors are characterized as relevantly similar to us, having our actual desires, motivations and conceptions of the good but just idealized in terms of their reasoning power and information, there is no reason to suspect their choice in representing our will. On the other hand, idealizations may be harmless in this way if we are aware of the differences between theoretical assumptions at hand and the actual situations.In applying idealized contract theories, we can always satisfactorily modify our judgments about actual persons like us by taking into account the differences between idealized beings and actual beings. This methodology is also practiced in scientific theories. [9] Idealizations are not irrelevant as long as the contractors are not characterized as beings extremely different from us.
I conclude that contract theories, if carefully refined and seriously defended, could rebut most of their alleged critics and challenges. I hope I have presented the whole idea well in this chapter to provide a useful context for later chapters. As I have argued above, hypothetical contract theories are initially plausible as a foundational theory. With this background, I will go though Scanlon's contractualist account of 'what we owe to each other', examining its own plausibility and weaknesses in Chapter 2. I will then discuss the general challenge posed by contractualism towards consequentialism, especially utilitarianism, in Chapter 3. In the succeeding chapter, I will concentrate on how contractualism can provide a plausible motivational basis for 'what we owe to each other'. If my defense for Scanlon's contractualism in these chapters is successful, a new picture on contract theories and normative ethics would in turn be drawn. This will be my focus of Chapter 5.
[1] Here, I am assuming fairness, adequate information and voluntariness as essential features that a morally binding contract should possess. It may be argued that it is circular or illegitimate to have such assumptions. I will argue why it is not necessarily circular or illegitimate when I discuss the critics of hypothetical contract theories.
[2] J. Habermas holds a similar view, see, for example, 'Discourse Ethics' in Moral Consensus and Communicative Action, (Cambridge, Mass.: MIT Press, 1990).
For a fuller account of how implicit contracts works, how they can lead to explicit contracts, and the relative strength of implicit contracts against hypothetical contracts, see Ronald J. Broach, 'Contractarianism in Ethics: Actual Contracts vs. Hypothetical Contracts', in Technology, Morality and Social Policy, ed. by Yeager Hudson, (Lewiston, NY : The Edwin Mellen Press, 1998) pp. 331-347.
[3]John Locke, Two Treatises of Government, ed. with an introduction and notes by Peter Laslett (Cambridge, Cambridge University Press, 1988), p.348.
[4] It may not be a problem faced solely by hypothetical contract theories, however. Actual contracts, either explicit or implicit, would also involve assumptions (e.g. depending on convention) in defining what situations constitute a contract. I have discussed this point in the critics of implicit theories. Hypothetical contract theories are particularly attacked because of their explicit assumptions in the contracting situation.
[5] This is a problem posed by Socrates in Plato's dialogue, Euthyphro. The original version is 'either what is pious is so because it pleases the gods or else it pleases the god because it is pious.'
[7] R. Dworkin, 'The Original Position', The University of Chicago Law Review, 1973 pp. 502-503, he uses it to argue against Rawls' employment of the original position in justifying his two principles of justice.
