Recent Criticisms of the Covenant of Works

in the Westminster Confession of Faith

© Cornelis P. Venema   (Vol. 9/3, Fall 1993, pp. 165-198, Mid-America Journal of Theology)   

"I. The distance between God and the creature is so great, that although reasonable creatures do owe obedience unto him as their Creator, yet they could never have any fruition of him as their blessedness and reward, but by some voluntary condescension on God's part, which he hath been pleased to express by way of covenant. II. The first covenant made with man was a covenant of works, wherein life was promised to Adam; and in him to his posterity, upon condition of perfect and personal obedience."  Westminster Confession of Faith, Chapter VII.

In his extensive study of the Westminster Confession of Faith (WCF), "The Assembly and Its Work," Benjamin Breckinridge Warfield remarked that "[t]he architectonic principle of the Westminster Confession is supplied by the schematization of the Federal theology, which had obtained by this time in Britain, as on the Continent, a dominant position as the most commodious mode of presenting the corpus of Reformed doctrine ...."[1]  Certainly, when the WCF is compared and contrasted with earlier Reformed confessions of the sixteenth century, it distinguishes itself by its full expression of federal or covenant theology, including this theology's characteristic distinction between the covenant of works and the covenant of grace.  No one reading the Westminster Confession of Faith can fail to detect the fruit of developments within Reformed theology on the doctrine of the covenant that occurred after the earliest period of Reformed theology in the first half of the sixteenth century.  For example, the WCF's distinction between a pre-fall covenant of works and a post-fall covenant of grace, a distinction which plays such a foundational role in covenant theology, is not found in the writings of John Calvin, and only first found expression among the Reformed in the writings of Zacharias Ursinus, an author of the Heidelberg Catechism.

The subject of the development of covenant theology in the period between the early Reformation and the writing of the Westminster Confession of Faith has been much discussed in recent literature.[2]  One of the disputed issues that has arisen in this connection is the degree to which the later, covenant theology is consistent with the earlier views of John Calvin.  Those who maintain a divergence of viewpoint between Calvin and the later covenant theologians frequently note that Calvin nowhere speaks of or develops a specific doctrine of a covenant of works.  Calvin, these writers repeatedly point out, only knew a covenant of grace.  It has also been argued that a significant divergence emerged within Reformed theology between, on the one hand, a "testamentary" or monopleuric view of the covenant shaped by the doctrine of election, and on the other hand, a full or dipleuric covenant doctrine.[3]  A great deal of ink has been spilt in evaluating these developments on the doctrine of the covenant in early and post-Reformation Reformed theology.

It is not my purpose in this article to review either the history of the development of covenant theology or to answer many of the questions that have arisen in the literature on this subject.  Rather, I want to consider some criticisms that have recently been registered against the formulation of the doctrine of the covenant of works in WCF.  Since the formulation of the doctrine of the covenant of works in the WCF expresses the dominant position of the covenant theology of the period in which the Confession was written (and indeed of subsequent Reformed covenant theology), these criticisms raise important questions regarding the warrant or biblical propriety of this doctrine.  Though it will not be my aim in what follows to set forth a full statement of this doctrine, I am interested in evaluating the validity of these criticisms and answering some of the objections that have been pressed against the WCF's understanding of the covenant of works.

It will become evident in what follows that there are two broad sources for such criticisms of the WCF.  The first arises primarily within the framework of neo-orthodoxy, that revision of classical Reformed theology associated with the theology of Karl Barth.  The second arises within the quite different framework of Reformed orthodoxy, though it represents something of an adjustment and refinement of the classical Reformed doctrine of the covenant of works.  After briefly summarizing the main lines of these criticisms of the WCF, I will conclude with a brief apologia in defense of the WCF's doctrine of the covenant of works.


Before considering two writers who have expressed the gist of the Barthian criticism of the WCF's view of the covenant of works, it will be helpful to consider a few important themes in the theology of Karl Barth which find their echo in this criticism.

Karl Barth

To understand Karl Barth's antipathy to the distinction between a pre-fall covenant of works and a post-fall covenant of grace, it is essential to grasp what he means by speaking of the creation as the "external basis of the covenant" and the covenant as the "internal basis of creation."  Speaking of the former, Barth argues that

[t]he existence and being of the creature willed and constituted by God are the object and to that extent the presupposition of His love.  Thus the covenant is the goal of creation and creation the way to the covenant.  Nor is creation the inner basis of the covenant.  ...  The inner basis of the covenant is simply the free love of God, or more precisely the eternal covenant which God has decreed in Himself as the covenant of the Father with His Son as the Lord and Bearer of human nature, and to that extent the Representative of all creation.  Creation is the external — and only the external — basis of the covenant.[4]

In Barth's theology of the covenant, God's free act of calling the creation into existence provides a context or setting for him to enter into covenant with the creature.  The creation constitutes the sphere within which God's gracious care and love for the creature in Jesus Christ can be expressed and realized.  It is in this sense, then, that the creation constitutes the external basis of the covenant, the arena within which God's saving purpose toward humanity in Christ can be realized and effected.

However, Barth also insists that the creation has no independent existence or meaning apart from the covenant of grace.  The covenant of grace, eternally purposed in Christ and realized in all of God's dealings with the creation, is the "internal basis of creation."

The covenant whose history had still to commence was the covenant which, as the goal appointed for creation and the creature, made creation necessary and possible, and determined and limited the creature.  If creation was the external basis of the covenant, the latter was the internal basis of the former.  If creation was the formal presupposition of the covenant, the latter was the material presupposition of the former.  If creation takes precedence historically, the covenant does so in substance.  If the proclamation and foundation of the covenant is the beginning of the history which commences after creation, the history of creation already contains, as the history of the being of all creatures, all the elements which will subsequently meet and be unified in this event and the whole series of events which follow; in the history of Israel, and finally and supremely in the history of the incarnation of the Son of God.[5]

The whole purpose of God's work of creation is the realization of communion and fellowship between God and his people in Christ.  In the free bestowal of his favor and mercy upon the creature in Christ, God shows his glory and realizes his purposes of self-revelation and self-communication to the creature.  In the covenant of grace, the triumphant "yes" of God to the creature of his favor resounds and the essential purpose of creation is realized.

In the development of Barth's theology of the covenant of grace, it is evident that he can find no place for a covenant of works, in distinction from the covenant of grace, that precedes in history the fall into sin and that does not express the saving grace exhibited in the gospel of redemption.  Not only does Barth regard the biblical account of creation and fall to be non-historical saga, but he also resists any suggestion of a transition in history from wrath to grace subsequent to the fall into sin.  From the beginning, God's dealings with the creature are pre-eminently and exclusively gracious.  There is no change that occurs in history in the relationship between God and the creature because of the fall into sin.  Furthermore, consistent with his view of the covenant of grace as the internal basis of creation, Barth rejects any ordering of law and gospel in which the gospel does not have the first (and as well, the last) word.  At not point in God's dealings with the creature does the law precede or antedate the gospel.  Not only in eternity, but also in history, the triumphant "yes" of God's grace has the first and definitive word.  To suggest that, prior to God's gracious dealings with his covenant people in the history of redemption, there existed another covenant relationship, a covenant of works, is to introduce a concept that betrays the most fundamental feature of all of God's dealings with humanity — the free turning of God toward humanity in Christ. 

From Barth's perspective, accordingly, a doctrine of the covenant of works like that enunciated in the WCF threatens the gospel of God's grace in Christ.  It rests upon a pre-critical view of the biblical history, viewing the biblical account of the creation and fall of man as though it were a straightforward account of historical occurrence and transition from favor to disfavor with God.  But more importantly, it permits the suggestion that humanity's covenant relationship with God, prior to the fall, might be construed as one based upon or at least contingent upon obedience to a probationary command and law of God.  The latter idea would entail placing law before grace in God's dealings with humanity before the fall into sin.  It would suggest that man's relationship to God, at least in the primal circumstances that obtained before the fall into sin, was found upon and sustained by meritorious works done in obedience to the law!

Many of the objections that have been offered to the WCF's understanding of the covenant of works stem from the influence of these themes in Barth's theology.  Though there are other sources for similar criticisms, it is the theology of Barth that informs many of the arguments against the legalism of the WCF's doctrine of the covenant, especially its formulation of the covenant of works.  To illustrate the influence of Barth and the nature of this criticism, it will be useful to turn to the arguments of two critics of the WCF that follow this approach.

Holmes Rolston III

One of the most vigorous advocates of the criticism that the WCF's doctrine of the covenant of works leads to legalism, is Holmes Rolston III.[6]  Rolston believes that the WCF's doctrine of a covenant of works represents a substantial betrayal of the original Reformation insight that man's standing before God is always founded upon grace alone.  When the WCF describes the covenant of works as a covenant in which "life was promised to Adam; and in him to his posterity, upon condition of perfect and personal obedience," it introduces into Reformed theology a concept of merit that militates against the genius of the Reformation rediscovery of the gospel of grace.

In Rolston's summarization of the classic view of the covenant of works, he maintains that it begins with and always insists upon the "merit and ability of man."[7]  In this first covenant, the Mosaic law's teaching that the promise of life is conditional upon man's obedience to the law (cf. Lev. 18:5) is read back into the original state described in Genesis 1 and 2.  The condition of this covenant is not faith, but works, and the reward of life is earned by law-keeping.  Thus, man's standing before God, his covenant fellowship with God, is founded upon and maintained by meritorious good works.  Furthermore, there is a tendency in this older covenant theology, Rolston insists, to identify the obligation of obedience with the natural law which binds man's conscience perpetually as a creature and which is to be sharply distinguished from the sphere of God's grace toward his people in Christ.  The extent to which this doctrine of the covenant is ruled by the idea of merit is evident also in its insistence that the saving work of Christ, the Mediator, involves a work of obedience, as the second Adam, in which the law is fulfilled on our behalf.  This is a doctrine, accordingly, that is wholly colored by the themes of obedience and merit and that mutes the gospel testimony of God's prevenient grace in all of his dealings with his people.

Rolston finds all of this in marked contrast with the order of grace that predominates and pervades the theology of John Calvin, and from which the WCF and the covenant theology it expresses is an obvious declension.  Calvin knew nothing, Rolston insists, of the two covenant doctrine of the WCF; in fact, he was not a covenant theologian at all, at least not in the normal usage of these terms.  For Calvin, "[a]ll things are ordered according to the movement of God's grace in creation and purpose in redemption."[8]  All of God's dealings with the creature, whether before or after the fall, are expressive of this order of grace.

Although Calvin does not use just that term, he speaks often of both the order and of the divine grace first instituted.  The part given to man is reflexive of grace.  From the start Calvin transcends the concept of order as primarily moral and legal and places this under the higher order of grace.  What is paramount is that God is gracious and requires acknowledgement of his grace.[9]

Calvin does not speak, therefore, of two covenants, a covenant of works and a covenant of grace, but of one order of grace, an order that may be either "inverted" through the fall into sin or                  "re-established" through redemption.  There are not two distinct covenants but one, the covenant of grace being a "reflection of and ... restoration of ... the original order."[10] 

There is no thought in Calvin's theology that man's relationship before God is sustained or maintained on the basis of meritorious good works.  For Calvin, grace always precedes the law, even in paradise, and man's obedience never merits God's acceptance but only expresses man's grateful and responsible answer to God's gracious dealings with him.  Life is always God's gift, never the achievement of the obedient creature.  What man ought to do is always reflexive of grace, unlike in later covenant theology wherein what man ought to do is fundamentally reflexive of law.  This also accounts for the superficial doctrine of sin in covenant theology.  Whereas covenant theology identifies sin primarily with disobedience to the law, Calvin identified sin with "man's faithless rejection of the goodness of God in favour of his own self-willed efforts to seek his own happiness elsewhere."[11]  Rolston finds the WCF, therefore, to be a serious departure, in its covenant doctrine, from the theology of John Calvin and the Reformers.  In this doctrine the grace of God in Christ is no longer the first or primary word.  In its place has come an emphasis upon man's legal obligation to his Creator by virtue of the covenant of works.

James B. Torrance

A second critic of the WCF who follows a similar line to that found in Barth is James B. Torrance.[12]  Torrance also regards the federal theology of the seventeenth century, especially as this is set forth in the WCF, to be a "rationalistic" departure from the early Scottish tradition of Knox, the Scots Confession, the pre-Westminster confessions, and the theology of John Calvin.  It is evident that Torrance believes the source for a growing legalism in Scottish theology and practice, confirmed in the so-called "Marrow Controversy,"[13] lay in an increasing emphasis upon the federal scheme and the conditional character of the covenant between God and his people.  The idea of "conditional grace" was introduced into Scottish theology, according to Torrance, by means of the route of federal theology.

Whereas the original, biblical idea of covenant expresses an unconditional binding of two parties in covenant loyalty and faithfulness, the federal theology shifted the emphasis from this notion of covenant to that of a legal contract.  Whereas a covenant is rooted in mutual promises and commitments, freely given and received, the federal theology, by distinguishing between a covenant of works and a covenant of grace, reconfigured the covenant as a contractual relationship.  This theology "is built upon a deep-seated confusion between a covenant and a contract, a failure to recognise that the God and Father of our Lord Jesus Christ is a Covenant-God and not a contract-God."[14]  In this understanding God appears as a Sovereign Employer and man as a servant-vassal.  No longer is the covenant relationship rooted in the gracious condescension of God to the creature, in which the creature stands as a beloved child or graciously-embraced friend.  Everything in the covenant of works has the color of a contractual relationship between Employer and employee, master and servant.

Torrance finds that a number of deleterious consequences follow from this reconceptualizing of the covenant in federal theology.  In the federal theology, there is the re-emergence, for example, of the older, Medieval view that grace presupposes nature and grace perfects nature.[15]  Contrary to the Reformation insight that nothing precedes God's gracious turning to the creature, this theology treats grace as a remedial measure, secondary to the original circumstance in which man stood before God under the obligation of the law of nature in a covenant of nature.  The priority of God's grace in all of his dealings with the creature is thereby imperiled.  Furthermore, in a criticism especially reminiscent of the theology of Karl Barth, Torrance regards the doctrine of limited atonement to be an extension of this covenant scheme.  The federal theology knows only one solidarity of all men and that is their solidarity with the first Adam whose fall into sin alienated the whole human race from God.  It does not know of that fundamental solidarity taught in the Scriptures in which all men are united in Christ, whose Headship extends over all creation and whose solidarity is inclusive of all men as Head of the human race.  This means that you cannot say to all men, "Christ died for you," though you may and even must say to all men, "You are all guilty and under judgment."  But according to Torrance, this betrays the triumphant note of joy that must resound in all gospel preaching — "you are what you are by God's grace in Christ!"[16]  All of this moves the accent from what God has done for us already in Christ to what we have to do for ourselves, if we would benefit from his saving work.

For these and other reasons, Torrance insists that the federal theology of the WCF is the primary culprit in stimulating, even reintroducing, a doctrine of meritorious good works into the room of Reformed theology.  The grand themes of the Reformation, sola gratia, solo Christo, sola fide, find in this theology their denouement. 

S. G. De Graaf and G. C. Berkouwer

The two previous critics of the federal theology which we have considered generally express a Barthian rejection of any distinction between a pre-fall covenant of works and a post-fall covenant of grace.  In their theological conception, there is ultimately but one order of grace that defines all of the Triune God's dealings with his covenant creature, whether before or after the fall (assuming that this distinction has historical significance).  This is not true, however, of the position of S. G. De Graaf and G. C. Berkouwer, two Dutch theologians who criticize the WCF's formulation of a covenant of works, but nonetheless admit a pre-fall covenant of "creation" or of "favor," as they prefer to term it, that must be carefully distinguished from the post-fall covenant of grace.  Consequently, though they criticize the WCF's formulation for its alleged "legalistic" implication, they do not deny the fundamentals of covenant theology or the distinction between two covenants, the one before, the other after the fall into sin.

The concern expressed by De Graaf and Berkouwer is that "[m]an's original life under God's rule cannot be regarded, for even a moment, apart from God's love and communion."[17]  The covenant of works formulation suggests that there is a legal order that obtains above or before the order of grace in God's dealings with his covenant creature.  In this understanding, the law and its obligation are easily separated from the life and fellowship with God that precede or antedate it.  Thus, the relationship between God and his covenant partner is misconstrued in the original circumstance before the fall into sin; it becomes the relationship between a servant and his Master, not that, for instance, between a child and his father or between a wife and her husband.  The bond of communion and fellowship, within which the law has its rightful place as a rule of life, is made to be secondary to the obligation or demand of obedience, on the basis of which God's favor is obtained.  As Berkouwer puts it,

[w]e err if we interpret this distinction as though God's original covenant had to do with our work or our achievement or our fulfillment of his law, while the later covenant of grace has reference to the pure gift of his mercy apart from all our works.  If we assume this we are compelled to say that God's original relation to man was strictly "legal," or that the structure of that relation was determined by man's merit.[18]

De Graaf likewise finds the idea of a covenant of works inimical to understanding the original covenant relation between God and man as one which was founded upon God's favor and goodness.  According to De Graaf, the idea of a covenant of works suggests that God's favor comes at the end or terminus of man's relation to God rather than at the beginning, as its source and foundation.  This intimates that the covenant life that man enjoys in fellowship with God comes only as a reward for obedience, as something merited or bestowed because man has made himself worthy of it.  Contrary to this suggestion, De Graaf maintains that

[t]here is never any speaking of merit or reward in the covenant of God, even in the so-called covenant of works.  God in his covenant is always the first who gives his love.  Through his love he must teach us to love; and our love can never be anything other than a response to his love.  Through the law God rules the fellowship of love we have with him, a fellowship which has no norm in itself, but for which God himself has established a norm.  And so it is that we by our responsibility to the norm grow in the communion of the love of God.  Thus the law is covenant law.  In the place of a "covenant of works," then, it is better to speak of a "covenant of God's favor."[19]

In the terminology of a "covenant of works," there lurks the tendency to abstract the law from its setting within God's original favor and kindness to the creature, a favor which gives the creature his place as a child and friend of God, called to obedience within the communion he already enjoys with the Creator.  This conception thereby conjures up the possibility of an absolute antithesis between works and grace, merit and favor, in which man's standing before God is thought, at least in its original and primary form, to be founded upon meritorious works.  Indeed, Berkouwer even goes so far as to argue that this language raises once more the specter of meritorious good works within the orbit of Reformed doctrine, a specter that threatens the Reformation criticism of the Catholic doctrine of justification by grace.

Therefore whoever burdens the so-called "covenant of works" with the notion of achievement and presumes that we gain God's favor in that way, must endorse the idea of a "nomological" ur-existence of man and must cut asunder the law of God from the fellowship of God.  In that way he isolates and hypostasizes the law.  It is not clear how this infusion of meritum can leave room for a genuine criticism of Rome concerning the question of meritoriousness  of  works.[20]

What is clear from these criticisms of Berkouwer and De Graaf is that they are concerned about a formulation of the "covenant of works" that places man's standing before God, in the original circumstance before the fall into sin, upon the foundation of merit and reward.  This kind of formulation threatens not only to distinguish the pre-fall and the post-fall covenants, but ultimately to oppose them; the first would be founded upon a principle of works, the second would be founded upon a principle of grace.  Therefore, though they both acknowledge a real difference between these covenants — this is in part the reason for De Graaf's suggestion that we speak of a "covenant of favor" for the first covenant, and of a "covenant of grace" for the covenant after the fall into sin — they do not wish to acknowledge that man's covenant fellowship with God is ever founded upon something other than the love and goodness of God in granting it as his gift.[21]  In this respect, their criticism of the formulation of a "covenant of works," though bearing some similarity to that of Karl Barth and those who follow in a Barthian line, is not to be confused with it.

John Murray

The last critic of the WCF's formulation of the doctrine of the covenant of works whom I wish to consider is John Murray.  Murray, though a faithful exponent of the system of doctrine contained in the WCF, was perhaps more critical of this aspect of the WCF than he was of any other.  Based upon his own biblical-theological reflection, Murray offered several of what he believed were needed correctives to the traditional formulations of federal theology, including the classical form found in the WCF.

Murray's original objection to the idea of a covenant of works stems from his reformulation of the doctrine of the covenant.  According to Murray, "covenant" in the biblical writings always expresses a gracious disposition of God toward the partner with whom he covenants; the notion of a "covenant of works," accordingly, is contrary to the ordinary meaning of covenant in the Scriptures, at least when they speak of God's covenanting with man.  In an encyclopedia article in which he traced briefly the history of covenant theology, Murray initially voiced this reservation about the older federal theology's doctrine of a covenant of works.[22]  However, he provided a more complete statement of his revision of the doctrine of the covenant works in his important article, "The Adamic Administration."[23]   As the title of this article suggests, Murray objected to the language of a covenant of works, not only in that it militated against the gracious character of God's covenanting with man, but also in that it speaks of a pre-fall "covenant," whereas the Scriptures reserve the language of covenant to God's post-fall dealings with the sinful creature.

Murray opens his treatment of the Adamic administration, his preferred terminology for the pre-fall arrangement between God and Adam (as representative head of the human race), by noting that, prior to the special arrangement described in Genesis 2, man who was created in God's image existed in a relationship with God of "perfect legal reciprocity."[24]  By this language, Murray refers to man's original obligation to live before God in accordance with the demand of God's law, the demand that he love and serve God with all his heart, soul, strength and mind.  This obligation to love God, the original demand of God's law of nature, constitutes the perpetual obligation of man as a creature formed for free and responsible service to God, his Creator.  Were man to have fulfilled this law and preserved his original state of integrity, he would have continued righteous and holy before God, and "[i]n this righteousness he would be justified, that is, approved and accepted by God, and he would have life."[25]

However, in addition to this original circumstance of providence, a circumstance which Murray describes as "mutable" and absent "full-orbed communion with God in the assurance of permanent possession and increasing knowledge,"[26] Murray notes that the account in Genesis describes an additional "arrangement" or "administration" of God's providence, ordinarily termed the covenant of works.  In addition to the perpetual obligation of obedience under which Adam stood from creation, God also "gave to Adam a specific command or, more accurately, a specific prohibition."[27]  By means of a special prohibition (Gen. 2:17), to which was attached a particular threat of death, God entered into a peculiar relationship with Adam.  This relationship or administration threatened death and carried within itself the implicit promise of life, though this promise is only indirectly suggested by the reference in Genesis 3:22,24 to the "tree of life." 

The Adamic administration is, therefore, construed as an administration in which God, by a special act of providence, established for man the provision whereby he might pass from the status of contingency to one of confirmed and indefectible holiness and blessedness, that is, from posse peccare and posse non peccare to non posse peccare.  The way instituted was that of "an intensified and concentrated probation", the alternative issues being dependent upon the issues of obedience or disobedience (cf. G. Vos: Biblical Theology, 22f.).[28]

There are several respects in which Murray's treatment of this Adamic administration differs from traditional covenant theology.  As we have already noted, this difference is partially terminological; rather than speak of a "covenant of works" or a "covenant of life" (the language of the Westminster Shorter Catechism),[29] Murray prefers to speak of an "Adamic administration," noting that the language of covenant is not used explicitly in the Bible to describe this relationship.  But the divergence is far more than terminological.  Murray also avoids the terminology of "works" because he wants to underscore the fact that, though the relationship this arrangement establishes includes within itself a concentrated probation, this administration is "sovereignly dispensed by God," and is "not a contract or compact.  Sovereign disposition is its patent characteristic."[30]  Accordingly, this arrangement, though non-soteric or non-redemptive, is consonant with the essential characteristic of the biblical understanding of a covenant as a "sovereign, divine administration ... [that] continues without any modification or retraction of its benefits by the immutable promise and faithfulness of God."[31] 

The language, "covenant of works," fails to do justice to the "elements of grace entering into the administration."  Despite the fact that the condition of obedience is essential to the probation which comprises such an important component of the Adamic administration, Murray regards the gracious origin and sovereign disposition of this arrangement to be such as to prevent our legitimately terming it a "covenant of works."  By means of this Adamic administration, God promised Adam, were he to fulfill the terms of the probation, an entrance into immutable and perpetual life in communion with himself, a state of glory that would exceed the mutability and contingency of his original state.  This promise, according to Murray, is an instance of gracious condescension and kindness which God did not owe the creature, but which he was pleased to grant to him.  This promise would not be granted upon the principle of strict justice or merit — God's justice does not require that Adam should ever be granted the status of immutability in fellowship with God — but would be expression of God's undeserved favor.

Consistent with his aversion to the language of "covenant of works" and parallel insistence that God's grace and sovereign disposition are basic to the Adamic administration, Murray also challenged another commonplace of the older federal theology, namely, that the Mosaic economy or covenant included within itself a repetition of the obligation of obedience, first enunciated in the covenant of works.

The view that in the Mosaic covenant there was a repetition of the so-called covenant of works, current among covenant theologians, is a grave misconception and involves an erroneous construction of the Mosaic covenant, as well as fails to assess the uniqueness of the Adamic administration.  The Mosaic covenant was distinctly redemptive in character and was continuous with and extensive of the Abrahamic covenants.[32]

Apparently, because Murray wants to emphasize the gracious and sovereign disposition of the Adamic arrangement, as well as the essential graciousness of the biblical covenant of grace, he does not want to admit the legal requirement of obedience to be as integral to this arrangement or the post-fall covenant of grace, as was typically the case in the history of covenant theology.  Whereas the older covenant theology regarded this legal requirement to be integral to the covenant of works, and even to the covenant of grace, Murray wants to distinguish sharply between the natural obligation of obedience and the probationary obedience of the Adamic administration.  Thus, in treating the work of Christ as the second Adam, Murray also resists the usual understanding that it included, in an important sense, the fulfillment of the legal obedience required by the covenant of works.

The obedience Christ rendered fulfilled the obedience in which Adam failed.  It would not be correct to say, however, that Christ's obedience was the same in content or demand.  Christ was called on to obey in radically different conditions, and required to fulfill radically different demands.[33]

What you find in Murray's treatment of the WCF's doctrine of the covenant of works, then, is not so much a repudiation of any of its essential teaching as a revision and refinement of some aspects of the WCF's formulation that he finds objectionable or misleading.  Without denying the important sense in which Christ's mediatorial work involved an act of obedience as the second Adam, fulfilling Adam's original obligation of obedience, intensified and concentrated in the probationary command, Murray wants to accent the elements of grace in the "Adamic administration."  In Murray's judgment, the WCF's use of the common language of a "covenant of works" inadequately accounts for these aspects of the first covenant.  Furthermore, the WCF does not clearly indicate to the extent that it might have that this first covenant or "Adamic administration" was a divinely initiated and sovereignly administered disposition of God toward his image-bearers.


A careful evaluation of the preceding criticisms of the WCF's formulation of the doctrine of the covenant of works shows that there are several issues that are outstanding.  Particularly within the orbit of neo-orthodoxy and Barthian theology, the issue of the distinction between a pre-fall and a post-fall covenant is most pronounced.  Following the lead of Barth, critics of the WCF like Rolston and Torrance do not wish to distinguish between a covenant of works and a covenant of grace, since they regard this to be an illegitimate denial of the one order of grace which characterizes all of Triune God's dealings with his covenant creature.  Furthermore, though the other critics do not share the Barthian theologian's denial of a distinction between a pre-fall and post-fall covenant, they all register objections to the WCF's designation of the first covenant as a covenant of works.  It is frequently argued, as we have seen, that this terminology introduces the idea that man's standing before God, at least in the pre-fall circumstance, was founded upon meritorious good works.  This, it is argued, threatens to make the creature's fellowship and communion with God, not so much a gift of God's favor or grace, but a reward for good works.  It also threatens to so distinguish man's covenant communion with God before and after the fall that the essential meaning of the covenant relationship is altered; in the first instance, God becomes man's debtor, in the second instance man becomes God's debtor!  The issues that surface in these criticisms have to do not only with the terminology of a covenant of works, but the underlying doctrine of the covenant and the intrusion of the notion of merit into the relationship between creature and Creator.

Admittedly, these criticisms and the issues they raise cannot be answered satisfactorily in this article.  They call for a re-examination of the whole doctrine of the covenant, in biblical, historical and theological terms.  However, there are several things that can be said by way of defense of the WCF.  Without promising to resolve all of these issues or providing a complete answer to these questions, there are several considerations that need to be borne in mind, especially in order to avoid placing an unwarranted construction on the WCF's doctrine of the covenant of works.

Two covenants or one?

The first consideration has to do with the question whether we should confess, on the basis of Scriptural teaching, the reality of two covenants, one before and the other after the fall, or one.  Though this may appear to be an unnecessary question, since the answer may to some be so obvious, it is one which the foregoing account of recent criticisms of the WCF necessarily raises.

It has to be clearly understood that one of the most profound differences between the older covenant theology and much, though by no means all, modern theology, lies just at this point.  The theology of Karl Barth and his epigoni has ultimately no place for the biblical revelation of the history of creation, fall and redemption.  In the criticisms of Barth, Rolston and Torrance, the difference between man's situation before the face of God before and after the fall into sin is flattened out, even obliterated.  These theologians do not clearly echo the Scriptural teaching that man was originally created good, after God's own image, and placed in a covenant relationship of communion or fellowship with God.  This covenant communion with God, before the fall into sin, included promises and demands.  It was, moreover, a mutable relationship, liable to being broken and lost through sin and disobedience.  This original covenant communion, however the difference is articulated, cannot be identified with the post-fall situation in which God's covenant people through the covenant of grace are restored once more to communion with God.  This latter covenant, the covenant of grace, is a covenant with the new humanity in Christ, the Mediator, and involves the calling out of a people for God's own possession from among the whole, fallen human race.

When the WCF, therefore, speaks of a pre-fall covenant of works between the Triune God and all of humanity in Adam, and then distinguishes this pre-fall covenant from a post-fall covenant of grace, made with a new people, the new humanity, those who are in restored communion with God through Christ, it echoes the basic structure of the biblical story of creation, fall and subsequent redemption.  This structure of the biblical history, which constituted the fundamental given of the older covenant theology (and of orthodox Christian theology generally), is often missing in modern theological revisions of the covenant, of which Karl Barth's is a notable example.  In this revision there is no place any longer for a historical fall from favor with God through the sin and disobedience of our first parent and covenant representative, Adam.  Nor is there any place for a subsequent covenanting between God and his people in the covenant of grace, by means of which fallen man is restored to renewed covenant fellowship with God in Christ, the second Adam.  In this revisionist theology of the covenant, there is only one covenanting between God and the creature, a gracious covenanting in Christ, which spans — perhaps it would be better to say, which obliterates — the difference between man's state of original sinless integrity in communion with God and his subsequent re-introduction to communion in the covenant of grace.

Consequently, it is absolutely critical to a biblical theology of the covenant that we recognize the basic correctness of the WCF's distinction between man's fellowship with God before the fall and his renewed fellowship with God through the gracious work of Christ after the fall.  Even though some may choose not to speak of "covenant" in the pre-fall state (for example, Murray), this does not alter the fact that a theology faithful to biblical teaching must reckon with the difference in man's standing before God in the pre- and post-fall states.  This the WCF does by means of its formulation of a covenant of works and a covenant of grace.


A voluntary condescension

This brings us, however, to perhaps a more difficult point, and that has to do with the nature or character of the pre-fall relationship between the Triune Creator and his sinless image-bearers.  Here we have seen that the WCF has been frequently charged with a misconstrual of this relationship, a misconstrual that bases man's communion with God on the foundation of meritorious works.  Implicit in this criticism is the fear that the formulation of the WCF, though rightly distinguishing the pre- and post-fall states, so distinguishes man's communion with God before the fall from his communion with God by grace after the fall, that there is almost an antithesis in the meaning of covenant before and after the fall.  Or, to state it somewhat differently, the language of the WCF suggests a kind of equivocation on the meaning of covenant: in the one instance, it describes the relationship and communion of an Employer and employee, a Master and a servant (the pre-fall covenant of works), and in the other instance it describes the relationship of a Father and a child, or of a Husband and a wife (the post-fall covenant of grace).  The former covenant is merited, the latter is freely and graciously given.  The former covenant is a matter of justice, the latter is a matter of grace.

This construction of the meaning and significance of the WCF's doctrine of the covenant of works is open to serious objection, however.  It trades too much upon the explicit language of a covenant of works, and neglects to notice other aspects of the WCF's understanding of the covenant.  Specifically, it fails to note that the WCF also speaks of God's condescending favor in the covenant of works and of God's freely granted justice in the covenant of grace.  The full statement of the doctrine of the covenant in the WCF includes promise and demand as essential constituents of both the covenant of works and the covenant of grace.[34]

It is imperative to notice that the WCF, before defining the covenant of works in Chapter VII, II, begins with a statement of the way all of the Creator's dealings with the creature are ordered covenantally.  In this statement, we read that

[t]he distance between God and the creature is so great, that although reasonable creatures do owe obedience unto him as their Creator, yet they could never have any fruition of him as their blessedness and reward, but by some voluntary condescension on God's part, which he hath been pleased to express by way of covenant.

By employing the language of voluntary condescension, the WCF makes it clear that the original covenant relationship was a sovereignly administered bestowal of God's favor upon the creature.  In this covenant relationship, it is God who takes the initiative and condescends to the creature.  He enters into a communion with Adam as his "son" (Luke 3:38), in which a promise is made and an obligation stipulated.  This language, accordingly, expresses something of what Murray and others mean when they speak of the "gracious elements" in the covenant of works, or when they insist that this covenant was not based upon a principle of strict justice, namely, the principle that man receives from God in this covenant only that which he in the strictest sense deserves.  Adam is granted and established in a communion of life with his Creator in the covenant of works.  And though he is obligated by the terms of his probation to offer a free obedience to his Creator, this obligation does not stand at the forefront or as the foundation of the covenant relationship, but rather serves as the manner of its administration.[35]

This emphasis upon all of God's covenants as voluntary condescensions preserves, it seems to me, the WCF from the charge of depriving the original covenant of the element of God's favor and goodness, as though it were only a matter of strict justice between a Master and his servant.  Moreover, by its apparent distinction between the original natural state in which "reasonable creatures do owe obedience unto him [God] as their Creator" and the covenant of works, the WCF preserves the element of unmerited bestowal and grant in this original covenant.[36]  It simply cannot be argued convincingly that the WCF neglects this component of the original covenant relationship between God and the creature before the fall into sin and the institution of the covenant of grace.

The question of terminology

No consideration of the criticisms often brought against the WCF on the covenant of works can avoid dealing with the question of terminology.  Does the language of a covenant of works present, especially in view of what we argued in the preceding section, something of a one-sided understanding of the original covenant?  Perhaps it would be better to speak in terms of a "covenant of life," the language used for this covenant relationship in the Westminster Shorter Catechism.  This language, or language like that of "covenant of creation" or "covenant of favor," does not diminish or belie the fact that God's original covenant with man was a bestowal of his favor, an act of undeserved and sovereign kindness and goodness in which the Creator took man into communion with himself as a friend and child.  By contrast the language of a covenant of works seems to compel the conclusion that man's standing in this covenant was solely founded upon his own achievement or accomplishment.

One aspect of this question of terminology with which I am not directly concerned here has to do with the absence of the language of "covenant" in the Bible to describe the pre-fall state.  It is certainly true, as John Murray has argued, that the Scriptures do not describe the original relationship between God and his image-bearers as a "covenant," and therefore the use of this language is not expressly biblical.  However, the arguments in the history of theology for terming this relationship a "covenant" are, in my judgment, convincing.  These arguments openly acknowledge that the doctrine is not expressly set down in the Scriptures, but nonetheless there are a number of biblical teachings or sedes doctrinae that cumulatively warrant the designation of this relationship as a covenant, the normal biblical designation for God's communion with his redeemed people in the covenant of grace.[37]  When the constituent elements of a covenant are present, and when the normal biblical term for a divinely instituted and administered communion between God and his creature is "covenant," there should be no substantial objection to the usage of this language.  John Murray, who highlights the lack of an express biblical reference to this relationship as a communion, prefers the terminology of an "Adamic administration," but this terminology is not only alien to the biblical descriptions of the pre-fall state but also to the biblical descriptions of God's communion with man in general.

On the matter of terminology, there are two points that need to be made.  First, the terminology of "covenant of works" needs to be complemented by the alternative terminology of "covenant of life" or "covenant of favor" (or even "covenant of creation").  There is a one-sidedness in the language of a covenant of works that demands the usage of these alternative designations.  It is especially useful to speak of this covenant as a "covenant of favor" since this language reminds us of the fact that this covenant was initiated and established by God, placed man in an undeserved position of favor, and granted him life in communion with God which his obedience would maintain and unfold.  Even the obligation of obedience to the law, concentrated and intensified in the so-called probationary command, was an invitation to man to respond to his covenant Creator in heartfelt, thankful service.  These dimensions of the first covenant can easily be diminished, when the exclusive terminology for this covenant is that of "covenant of works."

However, there is also a second consideration which counter-balances this one.  The language of a "covenant of works" helps to emphasize what was integral to the first covenant, namely, the obligation of obedience on condition of which man would remain in covenant communion with God.  In the account in Genesis 2 of the stipulation of obedience, it is this dimension of the first covenant that is most prominent or outstanding.  The language of the WCF, accordingly, keeps clearly before us the fact that no communion with God is possible for man, certainly no communion in which man might enjoy the fullness of life, short of one in which he offers to his Creator a glad-hearted service or obedience.  Furthermore, this language clearly distinguishes the first covenant from the covenant of grace on precisely that matter which is most decisive.  Though the first covenant was indeed an undeserved bestowal of divine favor, it was a favor shown to a sinless creature who had not yet forfeited through sin any further claim upon God's goodness.  There is a real difference between undeserved favor shown a sinless, obedient creature, and the undeserved grace granted the disobedient covenant breaker. The language of the WCF helps to keep the difference between man's status before and after the fall clearly in perspective.  God's dealings with man before the fall were not gracious in the strict sense, at least not in the sense in which they were after the fall.  In the covenant of grace, the demand and obligation of obedience remains, but God graciously gives a Mediator through whom that demand and obligation are met.  What was promised man in the first covenant, on condition of "perfect and personal obedience," to use the language of the WCF, is given to the believer in the covenant of grace through the work of obedience of the second Adam.  Only through the free gift of that righteousness which belongs to Christ, the second Adam, does the believer who receives this gift by faith become acceptable to God and again find himself received into his favor as a child (Rom. 5:18-21).

This real difference between the first and second covenants is maintained by and reflected in the WCF's distinction between a covenant of works and a covenant of grace.  Though this language may not be complete or altogether satisfactory, it does well preserve the difference between a communion which, to be maintained and unfolded, requires free and heartfelt obedience, and a communion which, to be restored and regained, requires the gracious and merciful granting of eternal life through the work of a Savior.  It echoes the Scriptural truth that the life promised man in the first covenant is only restored to man and ultimately realized in eschatological glory in the covenant of grace.

What about "merit?"

In order not to leave one further loose end among the common objections to the WCF's doctrine of the covenant of works, it will be useful at this point to consider whether it is ever permissible to speak of "merit" in the relationship between God and his covenant creature.  One of the difficulties that emerges from a review of the criticisms of the WCF is the ambiguity and impreciseness of the usage of the language of "condition" and "merit" in many of the discussions.

It should be evident from the foregoing that there is one obvious sense in which the language of "merit" has no place in a biblical theology of the covenant.  At no point in God's dealings with man as covenant creature may we say that God, in the strict sense of justice, owes the creature anything.  Everything God bestows upon the creature, whether in creation or redemption, is an undeserved favor or benefit of his goodness and kindness.  This holds true as much for man in his original state as in his redeemed state.  Though there is a difference between the favor shown sinless man and the grace shown the sinner, the covenant breaker who has forfeited any claim upon God's continued favor or goodness, at no point in God's dealings with man may we say that man gets what he deserves from God.

However, the fact is that God has, by entering into covenant with man, bound himself by the promises and as well the demands/obligations of that covenant.  This means that Adam's obedience to the stipulated obedience, though it were an outworking and development within the covenant communion in which he was placed by God's prevenient favor, would nonetheless "merit" or "deserve" the reward of righteousness God himself had promised.  In the covenant itself God bound himself to grant, as in some sense a reward well-deserved, the fullness of covenant fellowship into which Adam was called.  The terms of the stipulation of obedience — the explicit threat of death in the case of disobedience, the implicit promise of life in the case of obedience — warrant a qualified use of the language of "merit" or "reward."

This becomes especially significant, when we consider the work of obedience of Christ, the covenant Mediator and second Adam.  Christ, by his obedient fulfillment of all that which the law required, can legitimately be said to have merited or earned the Father's favor toward his people, those for whom he actively obeyed the law and on whose behalf he suffered its curse (his so-called "active" and "passive" obedience).  In so doing, Christ met the conditions of the first covenant and obtained for his people a favor previously lost through the disobedience and sin of their first federal head, Adam.

It is interesting, in this connection, to notice that Calvin, to whom many appeal in their criticism of the WCF's use of the language of "works" and of "condition" in describing the first covenant, explicitly defends the practice of speaking of Christ's work as meritorious!  Admittedly, Calvin readily acknowledges that Christ's work is not meritorious in the sense that our salvation finds its ultimate source in God's justice.  Christ himself, in his Person and work as Mediator, is wholly the gracious gift of the Father on behalf of his people.  In that sense, all of Christ's work finds its source in the grace, the unmerited favor, of God.  Nonetheless, integral to the gracious work of Christ is an obedience, after the pattern of Adam's disobedience, which remedies our circumstance as sinners by meriting God's favor and restoring us to a state of acceptance with God.[38]  Christ's work as Mediator, including both his perfect obedience to the demands of God's law and his suffering of the penalty due the law-breaker, merits or earns salvation for his people.  Though all stems from God's undeserved grace or favor, this does not mean that the actual work accomplished by Christ in no respect can be described as "meriting" salvation for his people.

Thus, whether one is fully satisfied with the WCF's choice of words when it speaks of a "covenant of works," it remains true that the biblical teaching which the covenant doctrine of this confession expresses is the common inheritance of the Reformed churches historically.  This inheritance has always understood the work of Christ, in the context of redemption and the covenant of grace, to be one which restores fallen man to that original favor and communion with God in and for which he was first created.  The covenant of grace is a post-fall remedy for the rupture in the covenant relationship between God and man brought about by the failure of Adam to live happily in terms of the first covenant.  And it is a remedy that fulfills the covenant creature's obligations to his Creator, thereby restoring him to fellowship with God.


Though there are questions unanswered and problems unresolved in the preceding defense against many recent criticisms of the WCF's doctrine of the covenant of works, the foregoing should sound a note of caution against jettisoning its formulations too quickly.  It is especially important that critics of the WCF, especially those who write from within the framework of a commitment to historic Reformed orthodoxy, not unwittingly join their voices to those who do not share this commitment and whose criticisms arise out of a radically unbiblical framework.

The WCF's formulation of the doctrine of the covenant of works rightly preserves the difference between the covenant of favor which man enjoyed before the fall into sin and the covenant of grace by which this favor, once forfeited, is restored.  It preserves the real historical difference between sinless man's fellowship and communion with God, a fellowship to be expressed and fulfilled in the way of obedience, and the sinner's restoration to fellowship through the work of Another, the second Adam.  Furthermore, the WCF reminds us in its covenant doctrine that the saving work of Christ, the Mediator of the covenant of grace, involved not only an atoning death which satisfied the truth and justice of the first covenant, but also a saving life of obedience by which man's creaturely/covenantal obligation to his Creator was fulfilled.  In so doing, the WCF helps us to see more clearly the glory of our covenant Mediator, by whose life, death and resurrection the believer is restored to covenantal life and fellowship with God.

    [1]The Westminster Assembly and Its Work (New York: Oxford University Press, 1931), 56.

    [2]Cf., Mark W. Karlberg, "The Mosaic Covenant and the Concept of Works in Reformed Hermeneutics: A Historical-Critical Analysis with Particular Attention to Early Covenant Eschatology" (Th.D. diss., Westminster Theological Seminary, 1980); idem, "Reformed Interpretation of the Mosaic Covenant," Westminster Theological Journal 43 (Fall, 1980), 1-57; idem, "The Original State of Adam: Tensions Within Reformed Theology," Evangelical Quarterly 87/4 (1987), 291-309; Ernest F. Kevan, The Grace of Law: A Study in Puritan Theology (1964; reprint Grand Rapids, MI: Baker Book House, 1976); Peter Alan Lillback, "Ursinus Development of the Covenant of Creation: A Debt to Melanchthon or Calvin," Westminster Theological Journal 43 (1981), 247-88; idem, "The Binding of God: Calvin's Role in the Development of Covenant Theology" (Th.D. diss., Westminster Theological Seminary, 1985); Michael McGiffert, "From Moses to Adam: the Making of the Covenant of Works," Sixteenth Century Journal 19/2 (1988), 131-55; Geerhardus Vos, "The Doctrine of the Covenant in Reformed Theology," in Redemptive History and Biblical Interpretation, ed. Richard B. Gaffin (Phillipsburg, NJ: Presbyterian & Reformed, 1980), 234-67; David A. Weir, The Origins of the Federal Theology in Sixteenth-Century Reformation Thought (Oxford: Clarendon Press, 1990).

    [3]Cf., Leonard Trinterud, "The Origins of Puritanism," Church History 20 (1951), 37-57; Richard Greaves, "The Origins and Early Development of English Covenant Thought," The Historian 21 (1968), 21-35; J. Wayne Baker, Heinrich Bullinger and the Covenant: The Other Reformed Tradition (Athens, OH: Ohio University Press, 1980).  This thesis of a twofold development in Reformed theology, or a divergence between two incompatible views of the covenant, has been subjected to vigorous criticism.  See Lyle D. Bierma, "Federal Theology in the Sixteenth Century: Two Traditions?," Westminster Theological Journal 45 (1983), 304-21; idem, "Covenant or Covenants in the Theology of Olevianus," Calvin Theological Journal 22 (1987), 228-50; idem, "The Role of Covenant Theology in Early Reformed Orthodoxy," Sixteenth Century Journal 21/3 (1990), 453-62.

    [4]Church Dogmatics, vol. III/1, The Doctrine of Creation (Edinburgh: T. & T. Clark, 1958), 97.

    [5]Church Dogmatics, III/1, 231-2.

    [6]Holmes Rolston III, John Calvin versus the Westminster Confession (Richmond, VA: John Knox, 1972); idem, "Responsible Man in Reformed Theology: Calvin Versus the Westminster Confession," Scottish Journal of Theology 23 (1970):129-56.  Rolston's argument is most succinctly stated in the second of these sources, from which I will draw primarily for my summary of his criticism.  The former source is an expanded version of the earlier article.  Rolston clearly writes from a revisionist perspective, even regarding the Confession of 1967 of the United Presbyterian Church to mark the end of Presbyterianism's venture in covenant theology.

    [7]Rolston, "Responsible Man in Reformed Theology," 133.

    [8]Rolston, "Responsible Man in Reformed Theology," 137.

    [9]Rolston, "Responsible Man in Reformed Theology," 139.

    [10]Rolston, "Responsible Man in Reformed Theology," 141.

    [11]Rolston, "Responsible Man in Reformed Theology," 150.

    [12]James B. Torrance, "Covenant or Contract? A Study of the Theological Background or Worship in Seventeenth-Century Scotland," Scottish Journal of Theology 23 (1970), 51-76; idem, "Calvin and Puritanism in England and Scotland — Some Basic Concepts in the Development of `Federal Theology,'" in Calvinus Reformator (Potchefstroom, South Africa: Potchefstroom University for Christian higher Education, 1982), 264-77; idem, "Strengths and Weaknesses of the Westminster Theology," in The Westminster Confession, ed. Alisdair Heron (Edinburgh: St. Andrews, 1982), 40-53.  In what follows I will trace Torrance's argument primarily as it is found in the first of these articles.

    [13]The "Marrow Controversy" refers to an ecclesiastical dispute within the Scottish Presbyterian church in the early eighteenth century.  In 1718 James Hog of Carnock republished The Marrow of Modern Divinity, thought to be a work of Edward Fisher, an English Calvinist of the seventeenth century.  This two-volume work criticized "neo-nomianism," the reintroduction of an inappropriate understanding of the believer's obligation to fulfill the law's demands in order to obtain or be maintained in the way of salvation.  Though the book pleased such notables as Thomas Boston, it met with considerable hostility within the Scottish Presbyterian church.  Torrance regards this chapter in the church's history to be instance of the growing influence of a legalism, earlier introduced by means of the WCF's doctrine of the covenant of works. 

    [14]Torrance, "Covenant or Contract?," 66.

    [15]Torrance, "Covenant or Contract?," 67.

    [16]Torrance, "Covenant or Contract," 69.  In this aspect of his argument, Torrance is also following Barth's lead in denying a transition from wrath to grace in history, subsequent to the fall into sin.

    [17]G. C. Berkouwer, Sin (Grand Rapids, MI: Eerdmans, 1971), 206.

    [18]Berkouwer, Sin, 207.

    [19]Het Ware Geloof, pp. 31-32 (translation mine).  De Graaf makes the same point, suggesting that we speak of a "covenant of favor," in his article, "De Genade Gods en de Structuur Der Gansche Schepping," Philosophia Reformata 1, 20-21: "Het is daarom m.i. beter, niet meer te spreken van `werken verbond', waardoor we onwillekeurig de gedachte aan verdienen en loon, en dan ook aan een voorloopig Zich terughouden van God in het Paradijs indragen. De term `verbond van Gods gunst' drukt beter uit de wekelijke verhouding.  Voor het verbond na den zondeval blijft dan de term `verbond van Gods genade', dat is van Gods schuldvergevende gunst."

    [20]Berkouwer, Sin, 208.  Cf. C. Van der Waal, The Covenantal Gospel (Neerlandia, Alberta: Inheritance Publications, 1990), 47-56.  Van der Waal severely criticizes the doctrine of a covenant of works for similar reasons.  He argues that Adam "was not created to be a legitimate pharisee, pelagian, or remonstrant" (54), as the doctrine of a covenant of works suggests.

    [21]See Daniel P. Fuller, Gospel and Law: Contrast or Continuum: The Hermeneutics of Dispensationalism and Covenant Theology (Grand Rapids, MI: Eerdmans, 1980), 18-64; idem, The Unity of the Bible (Grand Rapids, MI: Zondervan, 1992), who argues that covenant theology shares with dispensationalism an unfortunate and unbiblical disjoining of grace and law, as well as a disordering of the relation as one of law first and then grace.  Though I will address this criticism in the second part of this article, Fuller's criticism, like that of Barth and others, fails to do justice to the biblical history of creation, fall and redemption, treating the pre-fall circumstance of man as though it were in almost every particular the same as the post-fall circumstance.

    [22]John Murray, "Covenant Theology," in The Encyclopedia of Christianity (Marshallton: National Foundation for Christian Education, 1972), 199-216.

    [23]"The Adamic Administration," Collected Writings 2 (Carlisle, PA: The Banner of Truth Trust, 1977), 47-59.

    [24]Murray, "The Adamic Administration," 47.

    [25]Murray, "The Adamic Administration," 47.

    [26]Murray, "The Adamic Administration," 47.

    [27]Murray, "The Adamic Administration," 48.

    [28]Murray, "The Adamic Administration," 49.  Cf. Karlberg, "Reformed Interpretation of the Mosaic Covenant," 48-53, who criticizes Murray at this point for separating the obligation of obedience to the law, under which Adam stood at creation, from the peculiar obligations of this Adamic administration.  He suggests that thereby Murray separates from this "Adamic administration" or first "covenant" the obligations of obedience which Christ, the second Adam, fulfilled in the covenant of grace (Rom. 5:12-21).  Though Murray's formulations elsewhere suggest such a separation, it seems clear from this statement (with its quotation from Vos) that the special obligation of obedience in the Adamic arrangement is simply an intensification or concentration of the original obedience owed God by Adam by virtue of creation.  Thus, though it may be true that Murray treats the "Adamic arrangement" as a kind of addendum to the original state of creation, Murray does clear himself somewhat by this statement of the charge that he has posited a dichotomy between a pre-fall state of nature (the obedience to law required of man as creature) and of grace (the probationary command with its promise of eternal life as component of a gracious and sovereign administration, a peculiar providence).

    [29]The Westminster Shorter Catechism, Q. & A. 12, reads: "What special act of providence did God exercise towards man in the estate wherein he was created?  When God had created man, he entered into a covenant of life with him, upon condition of perfect obedience; forbidding him to eat of the tree of the knowledge of good and evil, upon pain of death."

    [30]Murray, "The Adamic Administration," 50.

    [31]The Covenant of Grace: A Biblico-Theological Study (London: The Tyndale Press, 1954), 14.  This language is taken from Murray's summary of what is essential to the biblical view of the covenant of grace.  It is striking to what extent Murray fashions his definition of the Adamic administration along lines that parallel his definition of the covenant of grace.  This allows him to show the similarity in the covenant relation before and after the fall.  However, because his definition of the covenant somewhat one-sidedly emphasizes sovereign promise, the obligation of obedience tends to be separated from or no longer integral to the covenant itself.  This is also an aspect of Karlberg's criticism of Murray's formulation: it does not permit the obedience of Christ, the second Adam, to be understood as a fulfillment of man's obligation under the covenant of works.  For Murray this obedience is integral to Christ's saving work, but it is so as a fulfillment of man's natural obligation of obedience, not directly as a fulfillment of that obedience required under the covenant of works.  Cf. Karlberg, "Reformed Interpretation of the Mosaic Covenant," 52-53; idem, "The Original State of Adam," 297-300.  Karlberg correctly identifies some of these problems in Murray's revision of the covenant doctrine, though he also is careful to note that they do not represent a repudiation of any essential doctrine confessed in the WCF.  

    [32]Murray, "The Adamic Administration," 50.

    [33]Murray, "The Adamic Administration," 58.

    [34]Meredith Kline, in his Treaty of the Great King: The Covenant Structure of Deuteronomy, Studies and Commentary (Grand Rapids, MI: Eerdmans, 1963) and By Oath Consigned: A Reinterpretation of the Covenant Signs of Circumcision and Baptism (Grand Rapids, MI: Eerdmans, 1968), argues for a distinction between "law covenant" and "promise covenant" that elucidates the difference between the covenant of works and the covenant of grace.  According to Kline, the covenant of works is the original and basic type of covenant administration in Scripture.  It is a law covenant in which obedience is stipulated, and a promise and sanction or threat are attached.  Such a law covenant operates according to a "law-inheritance" principle; obedience to what is stipulated "merits" the inheritance.  The inheritance is not "merited" in the strict sense of intrinsic merit; it is "merited" ex pacto, as the older covenant writers would say, or it is a covenanted meriting.  The covenant of grace, by comparison, continues to recognize the foundational place of law or the stipulation of obedience, but there is now added the promise that God will mercifully fulfill this stipulation through Christ.  The covenant of grace, accordingly, continues to uphold the "law-inheritance" principle, so powerfully enunciated in the covenant of works, but it does so by way of the addition of an alternative "faith-inheritance" principle, in which the covenant member receives life through faith in Christ, the Mediator of the covenant of grace.  Christ's obedience to the law and his suffering of the law's curse obtain life for the believer.  Though there is much in Kline's formulations that I find acceptable and praiseworthy, especially his insistence that the stipulation of obedience, first made in the covenant of works, remains operative in the covenant of grace, he tends to diminish the aspect of God's favor, as I prefer to speak of it, in God's original condescension to Adam, his image-bearer, in the covenant before the fall into sin.  Furthermore, his position suggests that Adam was not given life in communion with God from the beginning, but would only receive life later on, that is, not until or unless he fulfilled the principle of works-inheritance. 

    [35]Cf. De Graaf, "De Genade Gods en de Structuur der Ganse Schepping," who argues that, in the covenant of works, God's favor stands at the end rather than at the beginning of the covenant relationship.  This is not necessarily the case, at least not in the statement of the doctrine in the WCF.  Though it is true that the promise of life is upon condition of obedience, it is not true that Adam possessed nothing of that life, as a bestowal of God's condescending favor, at the beginning of the covenant relationship.  Part of the difficulty here is that Murray, following a hallowed tradition of Reformed theology, regards the promise of life in the covenant of works to be oriented to an eventual transition in Adam's standing before God from a state of mutability and contingency to one of irrevocable life in communion with God.  Because the obligation of obedience is regarded as a probationary obligation, it presumes some point of termination, at which time the promise of life would be irrevocably fulfilled.  This latter promise, though conditioned upon Adam's obedience to his probation, grants much more than strict justice would require and is, therefore, an instance or element of grace.  De Graaf makes the pertinent observation here, however, that God only threatens death in the way of disobedience.  The account in Genesis 2 does not teach that Adam was promised life only after, or at the successful completion of, a period of probation!

    [36]I have already mentioned in a preceding footnote Mark Karlberg's criticism of the WCF for introducing a "speculative element" with this distinction between man's natural state and the subsequent introduction of a covenant of works.  It is true that, in both the WCF and in the reformulation of John Murray, there seems to be implicitly present a distinction between the original state of nature and the subsequent covenant of works that parallels the older, Medieval distinction between nature and grace.  Since it is not clear that the covenant of works is original and native to man's circumstance as covenant creature, a state of nature is posited that antedates the state of "super-added" favor in the doctrine of the WCF and of John Murray.  In my judgment, this is more true of Murray's revision of the doctrine of the covenant of works than it is of the WCF.  The language of the WCF leaves some ambiguity here, but it could be read to teach that the nature of the difference between Creator and creature requires covenant as the medium of communion or fellowship.  If this is the case, then the WCF ultimately does not separate between a state of nature and a state of covenant before the fall, but suggests that the covenant of works is a kind of administration or particularizing of that covenant relationship in and for which man was originally created.  For a recent argument showing that man's original circumstance at creation was that of being in covenant with God, see: Mark Vander Hart, "Creation and Covenant Part One: A Survey of the Dominion Mandate in the Noahic and Abrahamic Covenants," Mid-America Journal of Theology 6/1 (1990), 3-18.  The parallels between the language used to describe the terms of man's original created state and calling, and that used to describe the re-establishment of man in communion with God in the covenant of grace, suggest that, biblically, man is be understood as from the beginning covenant creature.  

    [37]The only instance in which the Bible speaks of a "covenant" in connection with Adam is Hosea 6:7 ("But like Adam they have transgressed the covenant").  The meaning of the expression, _, has always been disputed, some taking it to be a reference to a place name, not Adam, the head of the human race.  However, the sedes doctrinae for the traditional description of the Creator's relationship with Adam before the fall are many.  Among them are the following: the explicit use of God's peculiar covenant name, ____, throughout Genesis 2 and 3; the presence of a variety of covenantal elements in the description of God's dealings with Adam in Genesis 2 and 3 (the sovereign administration of a peculiar bond or communion between the Lord and Adam, the stipulation of a particular obligation, the pronouncement of a sanction or curse, the implicit promise or "sacramental sign" of life in the "tree of life" mentioned in Genesis 3:22,24); the parallels in the language employed in Genesis to describe the pre-fall and post-fall relationship between God and his people; and the apparent reminiscences of the covenant of grace in its earlier administrations, as well as the pre-fall covenant communion of God with man, in the descriptions of the eschatological covenant communion in Revelation 21-22.  One hesitates here to apply the well-known words of Shakespeare, "a rose by any other name still smells like a rose," but it does seem appropriate.  Why not term something a "covenant," though the express term is not employed in the Bible, when the thing being described has all the earmarks of a covenant? 

    [38]Calvin: Institutes of the Christian Religion (ed. John T. McNeill, trans. Ford Lewis Battles; Philadelphia: The Westminster Press, 1960), III.xvii.1: "In discussing Christ's merit, we do not consider the beginning of merit to be in him, but we go back to God's ordinance, the first cause.  For God solely of his own good pleasure appointed him Mediator to obtain salvation for us.  Hence it is absurd to set Christ's merit against God's mercy.  For it is a common rule that a thing subordinate to another is not in conflict with it.  For this reason nothing hinders us from asserting that men are freely justified by God's mercy alone, and at the same time that Christ's merit, subordinate to God's mercy, also intervenes on our behalf.  Both God's free favor and Christ's obedience, each in its degree, are fitly opposed to our works.  Apart from God's good pleasure Christ could not merit anything; but did so because he had been appointed to appease God's wrath with his sacrifice, and to blot out our transgressions with his obedience.  To sum up: inasmuch as Christ's merit depends upon God's grace alone, which has ordained this manner of salvation for us, it is just as properly opposed to all human righteousness as God's grace is."  Though Calvin is speaking here in the context of redemption, and not creation, it is not difficult to see that his reasoning would equally well apply to the circumstance of Adam in the covenant of works.  Though Adam by virtue of God's favor and goodness was placed in fellowship with God from the beginning, his blessedness and continuance within this fellowship depended upon his grateful obedience to the stipulations of this communion.  In this latter, subordinate sense, you might say Adam would have "merited" or "deserved" the fellowship his obedience maintained.