THE INSEPARABLE BOND BETWEEN COVENANT AND

PREDESTINATION

COCCEIUS AND BARTH

Maarten Wisse

1. Introduction

Twentieth-century interest in federal theology in general and in

Cocceius’ theology in particular was primarily motivated by the inten-
tion to compensate for the post-Reformation Reformed interest in the
doctrine of predestination by the notion of the covenant. The alleged
role of predestination as the “Zentraldogma” gave so-called “Reformed
orthodoxy” the image of a harsh, rationalist, fatalistic system.1 In this
context, a strand of Reformed theology in which the loving fellowship
between God and believers played a crucial role was more than
welcome, fitting as it was into the typically twentieth-century interest in
thinking God as love.2 Thus, twentieth-century research on Cocceius
interpreted his theology as biblical rather than scholastic, historical
rather than rationalist, experiential rather than abstract.3

1 For the view of the doctrine of predestination as a “Zentraldogma,” see Willem J.

van Asselt, ed., Inleiding in de gereformeerde scholastiek (Zoetermeer: Boekencen-
trum, 1998), 18–30. Willem van Asselt has been one of the key critics of this so-called
“old school”-interpretation of Reformed scholasticism.
2 See Kevin J. Vanhoozer, “Introduction: The Love of God — Its Place, Meaning
and Function in Systematic Theology,” in Nothing Greater, Nothing Better: Theo-
logical Essays on the Love of God, ed. idem (Grand Rapids: Eerdmans, 2001), 1–29.
For two full-scale works on God as love, see Vincent Brümmer, The Model of Love: A
Study in Philosophical Theology (Cambridge: Cambridge University Press, 1993);
Markus Mühling, Gott ist Liebe: Studien zum Verständnis der Liebe als Modell des
trinitarischen Redens von Gott, Marburger theologische Studien 58, 2nd ed. (Marburg:
Elwert, 2005).
3 Willem J. van Asselt, The Federal Theology of Johannes Cocceius (1603–1669),
trans. from the Dutch by R. A. Blacketer, Studies in the History of Christian Thought
100 (Leiden: Brill, 2001), 2–16.
260 WISSE

Willem J. van Asselt is the present day expert on Cocceius.4 He

always resisted the oversimplified appropriations of Cocceius’ thought,
arguing that it is an anachronistic misreading of Cocceius’ work if one
contrasts it too much with the mainstream Reformed scholasticism of
his contemporaries.5
Still, van Asselt shares much of the twentieth-century worries about
the particularist aspects of Reformed theology. Two anecdotes may be
invoked to illustrate this. Once, I heard van Asselt reinterpret the tradi-
tional Dutch Reformed opening (votum) of a Church service. He para-
phrased “Who will never abandon the works of his own hands,”6 as
“Who will never abandon the work he has begun in each of us.” What
if his great seventeenth-century hero Johannes Cocceius had heard him,
who, as we will see, argues against this “heresy” to much length in the
Summa doctrinae! Also, when I was a student and expressed my worries
about the consequences of predestination thought to van Asselt, he
always replied with a quote from one of his teachers, the Dutch system-
atic theologian Arnold van Ruler: “The gospel skims across the border
of universalism.”7 Although van Asselt has been eager to criticize a
number of Karl Barth’s readings of the Reformed scholastics, when
confronted with the riddle of predestination, he often expressed his
sympathy with Barth’s universalisation of Reformed soteriology.
Given the combination of van Asselt’s expertise on Cocceius on the
one hand, and his appreciation for a Barthian solution to the problem of
predestination on the other, it seems appropriate to me to devote my
contribution to this Festschrift to the question of the relationship

4 Among his main works on Cocceius are a complete Dutch translation of Coc-
ceius’ Summa doctrinae: Johannes Coccejus, De Leer van het Verbond en het Testa-
ment van God, trans. from the Latin by W. J. van Asselt and H. G. Renger (Kampen:
De Groot Goudriaan, 1990) and, in addition to numerous articles, two monographs: a
more biographical one in Dutch (Willem J. van Asselt, Johannes Coccejus: Portret
van een zeventiende-eeuws theoloog op oude en nieuwe wegen, Kerkhistorische
monografieën 6 (Heerenveen: Groen, 1997)), and the thoroughly revised English trans-
lation of his dissertation: van Asselt, Federal Theology.
5 Van Asselt, Federal Theology, 94–105.
6 A traditional Dutch Reformed church service opens with the following phrase:
“Our help is in the name of the Lord, who has made heaven and earth, who will never
abandon the works of his hands,”—a combination of Pss 124:8 and 138:8.
7 In Dutch: “Het evangelie scheert langs de rand van de alverzoening.” More on
van Ruler in English: Allan J. Janssen, Kingdom, Office, and Church: A Study of A. A.
van Ruler’s Doctrine of Ecclesiastical Office, The Historical Series of the Reformed
Church in America (Grand Rapids: Eerdmans, 2006).
 INSEPARABLE BOND 261

between predestination and covenant theology in Cocceius and Barth.

On the one hand, my contribution presupposes and builds on van
Asselt’s Cocceius scholarship, his translation of Cocceius’ Summa
doctrinae and the two monographs. On the other hand, it poses a
friendly critique of van Asselt’s sympathy with a Barthian universalisa-
tion of Reformed soteriology. Before I turn to Cocceius, I introduce the
twentieth-century objections to Cocceius’ theology in a bit more detail
by outlining Karl Barth’s critique of federal theology. After having
introduced the reception of Cocceius in Barth, I will follow the main
steps of the development of the doctrine of the covenant in Cocceius’
Summa doctrinae (SD below),8 and confront Cocceius’ view of the rela-
tion between covenant and predestination with Barth’s universalisation
of election.
To issue a warning beforehand, I do not intend to develop a full-scale
theological defense of the traditional Reformed doctrine of predestina-
tion. Until recently, I myself accepted most of the common worries
concerning a theology of predestination. The present essay is the result
of a reassessment of these worries. These worries have not gone, but I
have now got an eye for some distinctive aspects of a traditional
Reformed soteriology that I had not noticed before. Thus, this contribu-
tion is intended to challenge the now commonplace Barthian view of
election as obviously theologically superior to everything that the
Reformed fathers had to offer.9 Instead, I aim to show that the theolo-
gical presuppositions and implications of Barth’s view precluded him
from taking some crucial soteriological notions properly into account in
the way the Reformed fathers like Cocceius were able to do. At the same
time, by way of a historical argument, my contribution confirms van
Asselt’s portrait of Cocceius as an “ordinary orthodox Reformed theo-
logian” rather than a “precursor of the Enlightenment” or “corrector of
Reformed predestinarianism,” as traditional scholarship had it. In fact,
I hope to put forward some of the good reasons Cocceius thought to
have to be as harsh a predestinarian thinker as his Reformed contempo-

8 The English translations of the quotations from the Summa doctrinae have been
prepared in close cooperation with drs. Jan Boom, who wrote his Master’s thesis under
the supervision of van Asselt on a Dutch translation of Aquinas’ and Cocceius’ com-
mentary on Lamentations 1. References to de SD are by chapter and paragraph number.
9 Bruce McCormack, “Grace and Being: The Role of God’s Gracious Election in
Karl Barth’s Theological Ontology,” in The Cambridge companion to Karl Barth, ed.
John Webster (Cambridge: Cambridge University Press, 2000), 97.
262 WISSE

raries, reasons that, to my conviction, have retained a good deal of their

validity over against the allegedly superior innovations of post-Barthian
twentieth-century theology.

2. Cocceius and the Barthian Tradition

In one of the footnotes to his magnum opus, The Federal Theology of

Johannes Cocceius, van Asselt characterizes Barth’s occupation with
Cocceius as follows:
In any case, Barth was occupied with Cocceius over the whole span of his
life—Cocceius caused this twentieth-century church father many a sleep-
less night! Often visitors would find Barth reading Cocceius. Through his
study of Cocceius the concept of the covenant became perpetually and
permanently conspicuous for Barth.10
Barth had three of Cocceius’ works in his personal library.11 In addi-
tion, his attention was drawn to Cocceius through his reading of
Schrenk’s book on federal theology.12 Barth’s evaluation of Cocceius’
doctrine of the abrogations depends almost entirely on Schrenk, given
that he probably did not have access to Cocceius’ main work on this
topic, the SD.13
In the Church Dogmatics, there is an extensive discussion of federal
theology in general, and Cocceius in particular, at pages 61–70 of
volume IV/1.14 In this excursus, Barth is very critical of federal
theology as a whole. It will be helpful to quote Barth’s general criticism
of federal theology at length, because it gives a good impression of the
main issue at stake in Barth’s relationship to federal theology:
But the more embracing and central and exact this apprehension becomes
in the main period of the Federal theology, the more insistently the ques-

10 Van Asselt, Federal Theology, 9.

11 Van Asselt, Johannes Coccejus, 107.
12 G. Schrenk, Gottesreich und Bund im älteren Protestantismus, vornehmlich bei
Johannes Coccejus (Gütersloh: Bertelsmann, 1923).
13 There is no single reference to Cocceius’ own works in the excursus on federal
theology. In volume II/2, where Barth appeals to Cocceius for his identification of
election and the covenant (See van Asselt, Federal Theology, 199–201 and van Asselt,
Johannes Coccejus, 222–225), he refers to the Summa Theologiae only: CD II/2, 85,
102, 114–5, 308; KD II/2, 91, 109, 122–3, 338.
14 CD IV/1, 54–66.
 INSEPARABLE BOND 263

tion imposes itself from what standpoint this occurrence is really

regarded and represented as such. What happens when the work, the
Word of God, is first isolated and then reconnected, according to the
teaching of pragmatic theology, with a whole series of events which are
purposefully strung out but which belong together? Does this really cor-
respond to the state of affairs as it is prescribed for theology in Scripture?
Can we historicise the activity and revelation of God? . . . They saw
excellently that the Bible tells us about an event. But they did not see that
in all its forms this narrative has the character of testimony, proclamation,
evangel, and that it has as its content and subject only a single event,
which in every form of the attestation, although they all relate to a whole,
is the single and complete decision on the part of God which as such calls
for a single and complete decision on the part of man. . . . The Federal
theologians did not notice that for all the exclusiveness with which they
read the Scriptures, in this analysis and synthesis of the occurrence
between God and man they were going beyond Scripture and missing its
real content. . . . As becomes increasingly plain in the sketches of the
Federal theologians, the atonement accomplished in Jesus Christ ceases
to be the history of the covenant, to which (in all the different forms of
expectation and recollection) the whole Bible bears witness and in face
of which theology must take up and maintain its standpoint, and it
becomes a biblical history, a state in the greater context of world-history,
before which, and after which, there are other similar stages.15
Most striking in this quotation is the overcritical attack on what Barth
calls the “historicizing” of theology that federal theology develops.

15 CD IV/1, 56–57. KD IV/1, 58–59: “Aber je umfassender, prinzipieller und

genauer diese Zusammenschau [vom Alten und Neuen Testament] in der Blütezeit der
Föderaltheologie wird, desto mehr drängt sich die Frage auf: von welchem Standort
aus dieses Geschehen nun eigentlich in Blick genommen und als solches dargestelt
sein möchte? Was geschieht da, wo das Werk, das Wort Gottes auseinandergelegt und
dann wieder pragmatisch-theologisch verknüpft wird zu einer Serie von sinnvoll anei-
nandergereihten und ineinander greifenden Ereignissen? Entspricht das wirklich der
der Theologie in der Schrift vorgegebenen Sache? Kann man Gottes handlung und
Offenbarung historisieren? . . . Daß die Bibel von einem Geschehen berichtet, das
haben sie vortrefflich verstanden, nicht aber, daß dieser Bericht in allen seinen Gestal-
ten den Charakter von Zeugnis, Verkündigung und Botschaft und zu seinem Inhalt und
Gegenstand ein einziges Geschehen hat, das je in dieser und dieser Gestalt seiner
Bezeugung, indem doch jede von ihnen sich auf seine Ganzheit bezieht, die eine, ganze
Entscheidung Gottes ist, die als solche nach der einen ganzen Entscheidung des Men-
schen ruft. . . . Die Föderaltheologen haben nicht bemerkt, daß sie zuerst mit ihrer Ana-
lyse und dann mit ihrer Synthese des Geschehens zwischen Gott und Mensch bei aller
Aufgeschlossenheit, in der sie die Schrift gelesen haben, an der wirklichen Schrift vor-
beilasen und an ihrem Inhalt vorbeisahen. . . . Ihm wird . . . die in Jesus Christus
geschehene Versöhnung aus der Bundesgeschichte . . . zu einer biblischen Geschichte,
zu einer Etappe in einem größeren Zusammenhang von Geschichte, vor der und nach
der es auch noch andere solche Etappen Gibt.”
264 WISSE

Even if Barth’s own theology is commonly presented as a Copernican

revolution in thinking about the relationship between God and history,
this quotation makes clear that the historicizing in Barth’s theology is of
a very special kind, namely a historicizing in the sense of the identifica-
tion of God with the human person Jesus Christ. If we had to believe
Barth, only a theology of the strictly Christological character that he
favours can do justice to the richness and complexity of the biblical
message! This is indeed the great divide between federal theology on
the one hand, and Barth on the other. Barth holds that theologies must
be based on and consist of only one thing: either Christ, as he claims his
own theology does, or sinful human nature, as he claims all theologies
accepting some form of natural theology do.16 As van Asselt has
pointed out, it is one of the central tenets of federal theology to think in
pairs, duplexities, as the English translator of his dissertation calls
them.17
In spite of the vigorous critique that Barth exercises in his excursus
on federal theology, there is much more positive influence of Cocceius
on Barth than Barth himself wants us to believe. In fact, throughout the
volumes of the CD, Barth is increasingly using covenantal conceptu-
ality to develop his theology. This starts in volume II/1, where Barth
introduces the notion of the covenant (as Gemeinschaft) in his doctrine
of God.18 According to Barth, God, by definition, is a God who chooses
himself to be a God in relation with human beings. This notion of the
covenant is then running through his doctrine of election and the divine
commandments in volume II/2,19 and it plays a central role in the
doctrine of creation in volume III. In volume IV, then, reconciliation
takes the form of the restoration of the covenant.20
In a way, Barth’s dogmatics can certainly be said to be a covenantal
theology. It is even appropriate to speak of a radicalization of federal
theology in Barth. Where Cocceius still sees the notion of the covenant
between God and human beings as characteristic of God’s works, not of
God’s being, at most metaphorically, this is different in Barth. Barth

16 On the Christological character of dogmatics, see CD I/2, 122–3; KD I/2, 134–

5. On the refutation of the knowledge of God from nature, see CD II/1, 63–127; KD II/
1, 68–141.
17 Van Asselt, Federal Theology, 303ff.
18 CD II/1, 272–296; KD II/1, 306–334
19 CD II/2, 3–33; KD II/2, 1–35.
20 CD IV/1, 1–78; KD IV/1, 1–82.
 INSEPARABLE BOND 265

sees the covenant as the one single definitive act of God’s being in Jesus
Christ. Exactly this is what accounts for the difference between Barth’s
Christological monism—although the term is overly pejorative—and
the duplex—not dualist as I hope to show—character of traditional
federal thought. For Barth, God’s relationship with human beings is
definitive of God’s being. In this sense, Barth holds that the covenant of
God is always a two-sided covenant, as God decided to be never without
a relationship to a human being—Jesus Christ. Still, Barth upholds the
one-sided origin of the covenant through an emphasis on God’s free
choice to be the way God is.
Hence, the “twentieth-century church father” cannot but vigorously
criticize a theology that speaks with two words rather than one. This is
especially true of two duplexities that are characteristic of federal
theology: first, the duplexity of the covenants of works and grace and,
second, the duplexity of the pactum salutis and the foedus gratiae.
As to the first duplexity, Barth’s conviction that God is God in Jesus
Christ makes it impossible for him to account for some sort of relation-
ship between God and human beings that is not a relationship mediated
by grace in Jesus Christ. For Barth, the idea of there being such a rela-
tionship suggests that we as human beings have some sort of natural
power to know God independently of God’s free decision to reveal
himself to us. If such a relationship not mediated by Christ is then also
the first, the original and the natural relationship between God and
human beings, even more natural than that through which God decided
to be the one decisive relationship with human beings, Barth can only
see that as an attempt to create one’s own god out of one’s sinful
mind.21 As to the second duplexity, the duplexity of the pactum salutis
and the foedus gratiae, Barth’s conviction that God is by definition God
in Jesus Christ makes it impossible for him to account for a level of
decision in God that is different from God’s definitive decision to be
God with us in Jesus Christ. For Barth, allowing for a pactum salutis
will inevitably lead to a dualism in God.22

21 CD IV/1, 56, 61–63; KD IV/1, 59, 64–66.

22 CD II/2, 94–116; KD II/2, 101–124.
266 WISSE

3. Cocceius on Pactum Salutis and Foedus Gratiae

The Covenant in General
In this essay, I will assess Barth’s twentieth-century critique of the
second abovementioned duplexity that is central to Cocceius’ theology,
the duplexity of the pactum salutis and the foedus gratiae. We start our
analysis with Cocceius’ definition of a covenant in general:
The covenant of God with a human being is different from the covenant
that human beings have among each other. A covenant between human
beings, namely, is based on mutual welldoing, whereas God makes a
covenant based on his welldoing only. The covenant of God is nothing
but the divine declaration concerning the way of receiving the love of
God, and of acquiring the union and communion with him. If a human
being makes use of this way, he is in a relationship of friendship with
God, or put differently: God is his creator and his God in a special way.23
One should notice that this is a definition of every covenant that God has
with human beings, no more, no less. As Cocceius points out, it is not a
definition of any covenant we can think of, because inter-human cove-
nants are different, as in inter-human covenants, both partners formulate
conditions and promises constitutive of the covenant. In a God-human
covenant, God’s declaration alone is constitutive of the nature of the
covenant.
So far, Barth and Cocceius are still on par. For Barth, it is very impor-
tant to stress that God is never dependent on the existence or actions of
human beings. Human beings only exist in the covenant with God by
virtue of God’s initiative in creation and revelation.
For Cocceius, this is as much the case as for Barth, although it is
significant that for Cocceius, this is still only the definition of the cove-
nant in general. That is, although both concrete covenants (of works and
of grace) between God and human beings are characterized by this defi-
nition, concrete covenants have some specific features that this defini-
tion does not contain. This is already hinting at the difference with
Barth. In a sense, for Barth, the general definition is a sufficient descrip-
tion of what God is for us in Jesus Christ. In Jesus Christ, God offers us

23 SD, I, 5: “Foedus Dei cum homine aliter se habet ac hominum inter ipsos.
Homines enim de mutuis beneficiis: Deus de suis foedus facit. Est enim Dei foedus
nihil aliud, quam divina declaratio de ratione percipiendi amoris Dei, & unione ac
communione ipsius potiendi. Qua ratione si homo utatur, in amicitia Dei est, sive, Cre-
ator ipsius est & Deus ipsius peculiari ratione.”
 INSEPARABLE BOND 267

a way in which we can enter into a relationship of love and friendship

with God. Human beings are called to respond to this offer in faith and
obedience. This offering of a way in which we can enter into a relation-
ship of love and friendship with God is constituted by God’s act of
reconciliation in Jesus Christ, in which God, acknowledging that we
will and cannot live in this friendship with God, responded to this call
in the ultimate way, suffering for us at the cross.24 As we will see, in
Cocceius, the consequences of sin make it impossible to account for the
specificity of the covenant of grace in terms of a mere offer of grace to
which we are called to respond.
Cocceius needs the distinction between a covenant in general and
specific covenants for two reasons: First, concerning the covenant of
works, Cocceius needs a specific covenant because the promise and the
conditions of the covenant of works are essential to the nature of this
covenant. Second, concerning the covenant of grace—more on that
below—Cocceius needs a specific covenant because the abrogation of
the covenant of works is of such a radical kind that a covenant in terms
of an obligation to be fulfilled on the part of human beings does not
suffice. After sin, human beings lack the ability to fulfil the obligations
of a covenant that asks something of them, if it not also “gives what it
asks”—Augustine.

The Covenant of Grace

The first concrete form that the covenant between God and human
beings takes is the covenant of works, but we skip the covenant of works
and its first abrogation in the fall for the moment. We will come back to
it in due course, and proceed to Cocceius’ definition of the covenant of
grace:
The covenant of grace is an agreement between God and a sinful human
being, in which, [first], God declares his free benevolence to give justice
and an inheritance to a certain seed in the Mediator through faith, to the
glory of his grace, [second], God invites through a commandment of
repentence and faith—or put differently: the repentence the beginning of
which is faith in the Mediator—, and through a promise, to give justice

24 As we will see below, here is the big tension in Barth’s conception. On the one
hand, God calls us to respond. Faith is exactly this response. On the other hand, only
God can respond to this call and does so in Jesus Christ, basically fulfilling the condi-
tion of the covenant for all human beings once and for all.
268 WISSE

to those who believe in him, [and third] the human being joins in the
agreed matters through cordial faith, resulting in peace and friendship
and the right to expect the inheritance with a good conscience.25
The most important difference between the general definition of a cove-
nant and the specific definition of the covenant of grace is the first part
of the latter—the declaration. In the first part of the definition, there is
no mention of a covenant between God and human beings, but of God’s
unconditional decree to save certain people through the mediatory work
of Jesus Christ: “God declares his free benevolence to give justice and
the inheritance of the covenant to a certain seed, to the glorification of
his grace.” Thus, in the definition, the decree is combined with the invi-
tation and the human response to the invitation, without the relationship
between the decree, the invitation, and the response being clarified.26
What is the background of this? The background is the first abroga-
tion of the covenant of works mentioned in the previous chapter of
Cocceius’ Summa Doctrinae: sin. According to the Reformed tradition
that Cocceius is following here, the power of sin is such that it makes a
natural human response to God’s invitation to the covenant of grace
impossible. If the covenant of grace is a mere proclamation of the work
of Christ for all humanity, leaving it to the responsibility of human
beings to accept this message in faith or not, no human being would be
saved, the Reformed fathers hold. Therefore, not a mere general procla-
mation of a common message is needed, but also the actual liberation
from the bondage of sin. This, then, is the reason why the covenant of
grace as an invitation to the love of God in Jesus Christ can only take

25 SD IV, 76: “Foedus gratiae est conventio inter Deum & hominem peccatorem,
Deo declarante liberum beneplacitum suum de justitia & haereditate certo semini
danda in Mediatore per fidem, ad gloriam gratiae ipsius, & per mandatum resipiscen-
tiae ac fidei sive resipiscentiae, cujus initium est fides in Mediatorem, ac per promis-
sionem justitiae credentibus in illo dandae invitante, homine autem per fidem cordis
astipulante contracta, ad pacem & amicitiam & jus expectandae haereditatis in bona
conscientia.”—emphasis mine. There are a number of subtle differences between van
Asselt’s translations of this definition and ours. Boom and I have read the definition as
built around the three verbs “declarare,” “invitare,” and “astipulare.”
26 Van Asselt has always insisted on the differences between the decree, the pac-
tum salutis, the testament and the covenant of grace: van Asselt, Federal Theology,
219–226, 239–247. Still, from a systematic point of view, it is important to see that
within the definition of the covenant of grace, reference is made to that which makes
this covenant possible, that is the eternal decree. This is not to suggest that the cove-
nant of grace (or parts of it) coincide with the eternal decree. Rather, I would say that
in the covenant of grace, the declaration of the eternal decree (in close relationship to
the pactum salutis), takes the form of a testament.
 INSEPARABLE BOND 269

effect if it is rooted in a testament. This testament is the declaration of a

pact between the divine Persons of the Trinity that guarantees the actual
salvation of certain people, whether these people accept it or not.27
Returning to Barth, this view of the implications of sin means that his
charge of natural theology in Reformed orthodoxy is unjustified. Given
that sin makes it impossible to know God without God’s actual inter-
vention in the life of human beings, there is no room for a human
attempt to reach God through the powers of one’s own autonomous
existence, as Barth fears. Hence, there are other ways to avoid the
dangers of an autonomous capturing of God than Barth’s option of
formulating the whole of Christian doctrine in Christological terms.
It may even turn out that the traditional Reformed distinction
between pre-fallen and fallen humanity provides a better way of
avoiding this trap than Barth’s Christological dogmatics. This,
however, depends on the specific strand in Barth’s thought that one
pursues. Following one line, there is ultimately only one true ontolo-
gical state of human beings, that is the state of being in relation to God
through Jesus Christ. Ultimately, human beings are what they are in
Jesus Christ even if they do not know or ignore it. Faith is not a change
of an ontological state. It is not a becoming of a new being in Christ. It
is just realizing what we have been all the time! This implies the risk of
“naturalizing grace” by accepting one’s relation to God in Christ as a
standing condition. It is something we can count on, whether we reckon
with it or not. Of course we need the revelation of God in Christ to know
in what state we are, but whatever we do in that state, the result is the
same. Barth, however, following another—I would say: opposing—
strand of his thought, would deny the possibility of “ontologizing” the
Christological foundation of his dogmatics. Being human in and
through Jesus Christ is never something one can count on, as it is always
a concrete gift of grace with a strict “here and now” character. Taking
our relationship with God in Christ for granted, Barth would say, is
precisely the proof of sin, as it takes us away from the dependence on
God’s free gift of grace.

27 Cf. van Asselt’s translation of the definition of the covenant of grace, who trans-
lates “conventio inter Deum & hominem peccatorem” as “an agreement between God
and sinful humanity”: van Asselt, Federal Theology, p. 41. This is incorrect, as the rest
of the definition shows. According to Cocceius, the covenant is only made between
God and the believer.
270 WISSE

Traditional Reformed orthodoxy accounts for the dependency on

grace in a different—I would say: more consistent—way. Here the
ontological state of fallen humanity is really distinct from the state
before the fall, and after the gift of grace. The gift of grace must be
understood as the beginning of the gradual restoration of one’s true
humanity to its pre-fallen state. In the state of the fall, the original image
of God is not entirely lost, but access to salvation requires a true onto-
logical change of what it means to be a human being.

The Pactum Salutis

Ontological change, however, is a delicate issue, as it is easily associ-
ated with coercion and determinism. It was not without reasons that
twentieth-century theology dropped the notion. Reformed theology
developed the duplexity of a “covenant from eternity,” the pactum
salutis, in order to provide the covenant of grace with a completely
secure foundation on the one hand, and maintain a sufficient degree of
human freedom on the other. Let us see how Cocceius develops it and
how he relates it to the covenant of grace.
The first thing significant to note about chapter V, the chapter in
which Cocceius extensively discusses the pactum salutis, is the title: “A
Further Explanation of the Foregoing.”28 This title is significant, as it
shows that Cocceius saw the pactum salutis not so much as a covenant
distinct from the covenant of grace, but rather as the eternal foundation
of the covenant of grace.29
The internal coherence of the pactum salutis and the foedus gratiae
is confirmed by the opening of chapter V of the SD, which is simply the
continuation of the previous line of argument:
However, in this divine testament [as discussed in the previous chapter,
MW], there is a pact that makes up its firmness. This pact, namely, is not
a pact with a fallen human being, but with the Mediator. This pact is the
will of the Father giving his Son to be a Head and Redemptor of the fore-
known people, and it is the will of the Son, setting himself up to take care

28 SD V, 88: “Uberior praemissorum explicatio.”

29 As such, it also appears in the STh where, contrary to the SD, the pactum salutis
precedes the foedus gratiae. In the STh, the pactum salutis denotes the eternal counsel
of salvation at the basis of the trinitarian work of salvation in time. The foedus gratiae
then, denotes the fulfillment of God’s counsel of salvation in a pact of friendship and
peace between God and the believer. STh XLI, 1–4.
 INSEPARABLE BOND 271

of this salvation. This will has the nature of an agreement insofar as, in
this ineffable economy of salvation, the Father is considered as the one
who stipulates the obedience of the Son to death, and as a reward for his
obedience, promises him a kingdom and a spiritual seed, and it is an
agreement insofar as the Son is considered as the one who sets himself up
to do the will of God, demanding the salvation of the people that were
given to him out of the world or, more clearly stated, claim his rights from
the other party.30
Several aspects of this quotation are worth noticing. First, the issue of
the strength (firmitas) of the testament. Why is the pactum salutis
needed to safeguard the firmness of the testament, and more generally,
of the covenant of grace? Should not God’s promise of salvation to all
who believe be firm enough? As we will see in more detail below, not
so for Cocceius. If the testament were only God’s promise of salvation
to those who believe, there would be no guarantee that the testament
would arrive at its destination at all. If the covenant were only an invi-
tation on God’s side, the sinner’s case would be hopeless, as the sinner
would be unable to fulfil the condition of access to the goods of the
covenant. Therefore, the covenant of grace, if it is to be a real answer to
the demand of human sinfulness, must include not only the invitation to
the friendship of God, but also the fulfillment of the condition of faith.
This is only possible if all conditions of the covenant of grace are met
in the trinitarian God, in the trinitarian pact. Therefore, secondly, it is
unavoidable that the covenant of grace as a whole, as regards its nature
as a testament, remains restricted to the elect, those “given to the Son by
the Father.”
Finally, it is significant that Cocceius speaks of the “ineffable
economy of grace.” The characterization of the pactum salutis as “inef-
fable” qualifies all contractual speech between the divine Persons, as
Cocceius explains in § 92:
Indeed, the will of the Father and the Son are the same, and not diverse,
because they are one. Still, insofar as the Father is not the Son, nor the

30 SD V, 88: “Inest tamen in hoc Testamento divino Pactum, quo nititur ejus firmi-
tas. Pactum scil. non cum homine lapso, sed cum Mediatore. Scilicet voluntas Patris
filium dantis caput & lutrwth&j redemptorem populi praecogniti, & voluntas Filii,
sese ad hanc salutem procurandam sistentis, habet rationem conventionis, dum secun-
dum ineffabilem illam oeconomiam negocii salutis notrae consideratur Pater stipulans
obedientiam Filii usque ad mortem, & pro ea ipsi regnum & semen spirituale repromit-
tens: filius autem se sistens, ad faciendam voluntatem Dei, & à Patre salutem populi
sibi è mundo dati restipulans, sive, ut claritis loquar, altrinsecus petens.”
272 WISSE

Son is the Father, this will is appropriated by each of both distinctly and
according to their own mode, to the one as sending and giving, to the
other as sent and given. Thus, this greatest mystery becomes known
(which had to become known to confirm our faith concerning our salva-
tion and to direct [this faith] to God), in what way we are justified and
saved by God, in what way God is, who both judges and vouches for us,
and is judged in that way, who absolves and intercedes, who sends and is
being sent.31
Cocceius’ insistence on the inexpressibility is significant vis-à-vis
Barth’s critique of the pactum salutis as a sort of contract between two
divine subjects, a view of the Trinity which is obviously incompatible
with Barth’s view of the trinitarian persons as modes of being.32 While
Barth refers to the Reformed tradition in support of his conception of the
trinitarian persons,33 the possibility of a pact between the trinitarian
persons in Cocceius makes clear how Barth’s conception differs from
the tradition. Whereas Barth’s modes of being in God are three ration-
ally conceived functions of a single subject, the traditional Reformed
view still conceives of the relationship between the one being of God in
three Persons as, well indeed: an ineffable relationship.34 In this inef-
fable relationship, indeed three more or less subject-like persons can be
distinguished, who at the same time, however, form an inexpressible
unity, both in themselves and in their works.

31 SD V, 92: “Patris quidem & Filii voluntas eadem est, non diversa, quia & unum
sunt; sed, quatenus Pater non est Filius, neque Filius Pater, eadem voluntas distincte &
suo modo utrique appropriatur, scilicet alteri ut donanti & mittenti, alteri ut dato &
misso. Ita mysterium illud maximum (quod fidei nostrae de salute nostra confirmandae
& in Deum dirigendae causa patescere debebat) patescit, quomodo in Deo justificemus
& salvemur, quomodo Deus sit & qui judicat & qui spondet, atque ita judicatur; qui
absolvit & qui intercedit; qui mittit & qui mittitur.”
32 CD IV/1, 64–65; KD IV/1, 68–69. See also van Asselt, Federal Theology, 233–
236. For some nuances concerning the use of the trinitarian persons as ‘modes of
being,’ see: Iain Taylor, “In Defence of Karl Barth’s Doctrine of the Trinity,” Interna-
tional Journal of Systematic Theology 5 (2003): 33–46.
33 See also CD I/1, 407–415; KD I/1, 374–381.
34 For a systematic account of the view of the Trinity as ineffable, drawing on
Augustine’s theology, see: Maarten Wisse, “‘Ego sum qui sum’: Die trinitarische
Essenz Gottes nach Augustins De Trinitate,” in Entzogenheit in Gott: Beiträge zur
Rede von der Verborgenheit der Trinität, ed. M. Mühling and M. Wendte, Ars Dispu-
tandi Supplement Series 2 (Utrecht: Ars Disputandi, 2005), URL: http://
adss.library.uu.nl, 63–76; idem, “De uniciteit van God en de relationaliteit van de
mens: De relevantie van Augustinus voor de hedendaagse theologie,” Nederlands The-
ologisch Tijdschrift 60:4 (2006): 310–328.
 INSEPARABLE BOND 273

Obviously, Cocceius’ conviction that the covenant of grace must

include a pact between the Father and the Son, has also significant rami-
fications for his view of the work of Christ. In § 104 the question is
raised for whom Christ has become Sponsor: for the elect only.
However, Cocceius is very careful with the use of predestination
language. He is always keen on explaining the soteriological context in
which the conclusion of limited atonement becomes unavoidable:
First of all, it is clear that, for whom he has vouched, for them he has also
succeeded, he has been their merit, their sins have been put on him and
they have been condemned in him, he has sacrificed himself for them and
has prayed for them; in addition, it is clear that those for which he died,
also died [in him]: that those are the same that have been justified and
saved through him. These things, namely, are of the same effect and
extent. . . . Since Scripture denies in the strongest wordings that the guar-
antee of Christ concerns all and each, no one excepted, and since it has
thus far been a generally accepted dogma in the Church that Christ, as it
has been said, did not die for all without exception according to the effi-
cacy [of his death], it can easily and safely be concluded (although it con-
cerns a great mystery), that Christ was no guarantee for all without
exception, or for those who are not saved.35
An extensive argument follows against the Arminians and Socinians,
who extended the benefit of Christ’s work to all people. What is at stake
in this argument time and again, is the content of what it means to say
that Christ died for someone. If, Cocceius argues, the scope of the
atonement in Christ is extended to all people, the material content of
what it means that Christ died for someone will change, and in
Cocceius’ view, it will loose its force. Thus, in § 113:
In no way should the phrase from Scripture be weakened that Christ has
died for sinners. This means much more than just that he has died to the
benefit of humans, insofar as at least is not meant that benefit that there is

35 SD V, 108: “Et primo quidem illud evidens est, pro quibus spopondit, illis &
impetravisse, illis meritum esse, illorum peccata in ipsum injecta suisse, & in ipso con-
demnata esse, pro illis se obtulisse, pro illis orasse; &, pro quibus mortuus est, illos
mortuos esse: eosdemque justificari & salvari per ipsum. Haec enim paris efficaciae &
ejusdem sunt latitudinis. . . . Quum igitur Scripturae apertissimis verbis negent, illam
sponsionem Christi ad omnes pertinere & singulos, nullo excepto, fueritque hactenus
in Ecclesia receptissimum dogma, Christum (ut loquuntur) secundum efficaciam non
esse mortuum pro omnibus hominibus sine exceptione: & facile & tutum est (licet in re
magni mysterii) definite, Christum non spopondisse pro omnibus sine exceptione, sive
etiam pro illis, qui non salvantur.”
274 WISSE

in the attainment of salvation, but in some other benefit, such as that they
are called, or that they are led to the knowledge of truth.36
And again in § 163:
There are others who reduce the merit of Christ, such as 1. Those who
state that Christ has died no more for those who are saved, as for those
who perish. Although they seem to extend the merit of Christ, in fact they
reduce it in such a way, that nothing remains of what he has merited.
Indeed they speak of the grace that is necessary and sufficient to believe
and to acquire reconciliation. But what is this [grace]? Is it the calling?
Impossible, for many are not called.37
So, if we bring this back to the discussion with Barth: for Cocceius, the
duplexity in the covenant of grace, that is the duplexity of the firmness
of the inter-trinitarian pact on the one hand, and the dynamics of invita-
tion and faith in time on the other, is absolutely necessary. If we, like
Barth, speak of only one decision in God, we will loose one of the two
elements: We will either loose the firmness, fruitfulness and effectivity
of God’s work of salvation, ending up in a theology in which God is in
some way dependent on human responsibility for salvation to come
about (Pelagianism/Arminianism), or we loose the dynamics of God’s
interaction with human beings in the preaching of the gospel, ending up
in hard universalism (unconditional salvation for all, regardless what
their response is).38 The problem of Barth’s position is that he refuses
to choose one of the two options.

36 SD V, 113: “Minimè enervanda est phrasis Scripturae, qua dicitur Christus pro
hominibus mortuus. Plus illud significat, quam mortuus utilitate hominum, siquidem
non utilitatem illam, quae est in assecutione salutis, sed utilitatem quamvis intelligas;
ut est, quod vocantur, quod ad agnitionem veritatis adducuntur . . .”
37 SD V, 163: “Sunt alii, qui imminuunt, videlicet 1. Qui statuunt Christum non
magis pro iis, qui salvantur, quam pro iis, qui pereunt, mortuum esse. Quanquam enim
videantur extendere meritum christi, reipsa tamen id adeò imminuunt, ut omnino nihl
ipsi relinquant, quod meritus sit. Dicunt quidem . . . Gratiam ad credendum & recon-
ciliationem consequendum necessariam & sufficientem. Quid illa? An vocatio? Non
potest. Plurimi enim non vocantur.”
38 I distinguish between “hard” and “soft” universalism. “Hard universalism” is a
view of salvation in which all will be saved, regardless of what their response is (the
so-called apokatastasis pantoon). “Soft universalism” is a view in which God promises
salvation to all, but makes it dependent on human decision whether it is actually real-
ized (popularly phrased: Arminianism).
 INSEPARABLE BOND 275

The Covenant of Grace as Communicative Act

So far, we have seen that for Cocceius, the covenant of grace needs to
be grounded in the pactum salutis—among other reasons—in order to
provide the covenant with the robustness required by the disastrous
effects of sin. This is the first central tenet on which the Reformed sote-
riology is built. Reformed theology would not be characterized by
duplexity, though, if there were not a second central notion constituting
it. As much as Reformed theology is concerned to maintain the firmness
of salvation, it is concerned to maintain the nature of salvation as a
communicative act. God saves by the Word, by proclaiming salvation
in Christ to human beings in the preaching of the gospel. Partaking in
salvation is a matter of a human act of response to the preaching of the
gospel.
But can these two notions live together in a peaceful way? The
charge of the Barthian tradition is that they cannot. In Barth’s view, the
Reformed view of Christ as the mirror of election cannot be consistently
thought together with a doctrine of double predestination, in which God
decides on the ultimate destination of human beings in an arbitrary way.
This is one of the main grounds for Barth’s reduction of the doctrine of
election to a communicative act: Election is the “Sum of the gospel.”39
According to Barth, the covenant between God and human beings can
only be a communicative act if there is no “secret decree” behind it.
Cocceius is of the exactly opposite opinion. He believes that the
communicative nature of the covenant of grace can only be truly safe-
guarded if it is rooted in the pactum salutis as God’s ultimate decision
on the destination of human beings that remains independent of the
communicative structure of the gospel. It needs to be independent of
this communicative structure because its firmness requires that it
remains independent of human consent.
The key passage in which Cocceius explains the inner logic of this
position is this:
This is of utmost importance to the foundation of faith and evangelical
consolation. And because God approves every truth that flows from his
counsel, one can rightly say that it is his will that everyone who sees the
Son and believes in him, has everlasting life. Although, namely, these
ALL are ONLY those given to Christ, and in God there is no universal
counsel without a determination of subject, or again, a decree to bless

39 CD II/2, 12–34; KD II/2, 11–35.

276 WISSE

without the explicit mentioning of a certain seed, nevertheless, through

his approving will, he wants to be universally true that which follows and
is implied by his special and definite counsel. . . . Through such a condi-
tional commandment and promise, salvation is offered to all those called,
i.e. it is proposed to them without any deceit; thus, it is clear that there is
no reason to suggest some sort of desire or incomplete will or the like that
God would be unworthy of, so that we uphold God’s integrity and sincer-
ity.40
This passage may require some explanation. Let me start at the end.
Cocceius’ emphasis on God’s integrity and sincerity can be technically
phrased as his conviction that the combination of a doctrine of predes-
tination (including limited atonement) with the free offer of Christ in the
gospel to all who hear it, is entirely consistent. No compromise of the
content of God’s eternal decree in the preaching of the gospel is
required, nor is the offer of Christ in the gospel to all in any sense an
insincere offer, a mere play to guarantee the responsibility of the non-
elect.
Cocceius provides the solution in the abovementioned key passage:
what God decides to work out from eternity is an unconditional
promise, taking the form of “God will do so and so whatever happens.”
At the same time, however, this decree to do so and so appears in the
preaching of the gospel in a conditional manner: “All those who believe
in Jesus Christ will be saved.” The latter is entirely consistent with the
former, as all those who believe in Jesus Christ will indeed be saved, the
eternal decree providing the certainty that those who receive the regen-
erating grace of God, will indeed believe in Jesus Christ. Thus, the
eternal decree of God in no way interferes with the free offer of Christ
in the gospel, because the believer-to-be does not in any sense need
access to the eternal decree in order to be allowed access to Jesus Christ
offered in the gospel. The Reformed theologians remain perfectly able

40 SD VI, 184: “Maximique id ipsum momenti est ad fundandam fidem & consola-
tionem Evangelicam. Et, quia Deus approbat omnem veritatem, quae ex consilio ipsus
fluit, rectè dicitur voluntas ipsius esse, ut omnu, qui videt filium & credit in ipsum,
habeat vitam aeternam. quanquam enim hi O M N E S sint S O L I dati Christo, & Deus
non habeat consilium universale sine determinatione subjecti, sive propositum benedi-
cendi citra vocationem seminis; tamen Voluntate approbante hoc vult universaliter
esse verum, quod ex speciali & definito ipsius consilio fluit & consequitur. . . . Per tale
mandatum & promissionem conditionatam omnibus vocatis salus offertur, h. e. pro-
ponitur sine omni illusione; ut patet neque necesse est singere desiderium sive vol-
untatem incompletam & alia istiusmodi Deo indecora, ut tueamur ipsius integritatem
& sinceritatem.”
 INSEPARABLE BOND 277

to quote Isa 55:1/Rev 22:17: “[W]hoever wishes, let him take the free
gift of the water of life.”41
Furthermore, the act of faith in Christ is and remains the sole point
of access to salvation. It is important to see that this is a crucial point of
agreement between the Reformed orthodox theologians and the Armin-
ians. Being saved is really about doing something, acting upon the
gospel proclaimed. The Reformed object against the Arminians’
unclarity about the origin of the act of faith, i.e., the question whether
and in what sense grace is necessary to make the act of faith possible,
but they do not dispute the character of faith as an act of response to
Christ offered in the gospel.42 If we put it in a popular way: What the
Reformed orthodox would have against the mass meetings of Billy
Graham is not the emphasis on making a decision for Christ. There is
much of such emphasis on making a decision in Reformed practical
literature, the Anglo-American Puritan tradition in particular. What the
Reformed tradition might have against a Billy Graham meeting is the
suggestion that one’s being able to make the right decision depends on
oneself rather than God alone. You may choose, but in choosing, the
only thing you can say is: “We love because he first loved us.” (1 John
4:19).
Of course, the conditional nature of the promise of salvation to all
who believe qualifies the object of the belief. What one has to believe is
not so much the fact that one is saved, but that those who believe will
be saved. This has important consequences for the question of assurance
of faith:
Question: is everyone in common obliged to believe that Christ has died
for them? Answer: This is exactly the consolation that is the fruit of jus-
tice; it pertains only to those who have a dismayed conscience, and to
those souls that hunger and thirst after justice. . . . Nobody may dare to
arrogate this consolation to himself who has not been converted to God
by true faith of his heart, i.e. who not hungers and thirsts after justice, and
[bears] fruits of that to the glory of God. Someone who has not taken ref-

41 See the earlier argument for this point in: Maarten Wisse, “‘Zij laat alles zoals
het is’: De actualiteit van de scholastieke methode,” in van Asselt, ed., Inleiding, 163–
173.
42 It must be said that there are some exceptions to this rule, compensating for the
negative consequences the emphasis on faith as an act might have in pastoral practice.
This compensation is particularly provided by the concept of faith as a habit. See
Maarten Wisse, “Habitus fidei: An Essay on the History of a Concept,” Scottish Jour-
nal of Theology 56:2 (2003): 172–189.
278 WISSE

uge in Christ, to put it concisely, who did not begin to love him as the
ruler of salvation.43
The position of the traditional Reformed theologians becomes all the
more clear when we confront it with Barth’s view. Barth’s single decree
of God to be God in Jesus Christ is motivated by his attempt to think
God exclusively as God with us, as God in relation to human beings.44
In addition, the attempt to think God as God in Christ exclusively is
motivated by Barth’s aim to dynamize the allegedly static under-
standing of God in the tradition.45 Barth’s aim is to bring history, the
contingent encounter between God and human beings in the here and
now, to the center of the theological discourse. Thus, for him, the
doctrine of election can be nothing but a form of communication, the
sum of the gospel.
However, as there is only room for one decree in God,46 and the
communicative message of the gospel cannot be the announcement of
those elected from eternity, Barth is forced to accept universalism.47
Thus, the message of the gospel can be nothing but an announcement of
a state of affairs, namely the state of being reconciled with God.
Although in Barth, God is defined by his being God in Christ in time,
the dynamics of God in time is in fact a dynamics of a single moment,

43 SD VI, 180: “Quaeritur, An omnibus omnino imperetur credere, Christum esse

pro se mortuum? Resp. Hanc ipsam esse consolationem, quae est fructus justitiae; &
non pertinere nisi ad conscientias contritas & animas esurientes & sitientes justitiae. . .
. Hanc consolationem nemo sibi debet arrogare, qui non vera animi fide conversus est
ad Deum; h. e. qui non sitit & esurit justitiam & fructus ejus ad gloriam Dei; qui non
confugit ad Christum, &, ut uno verbo dicam, qui non ipsum incepit amare, ut princi-
pem salutis,” Here, Cocceius is fully on par with Voetius: De scholastieke Voetius:
Een luisteroefening aan de hand van Voetius’ Disputationes Selectae, ed. W. J. van
Asselt and E. Dekker (Zoetermeer: Boekencentrum, 1995), 98–100.
44 CD IV/1, 1–22; KD IV/1, 1–22. Of course this is not to deny Barth’s emphasis
on the freedom of God to be God with us.
45 CD II/1, 257–271; KD II/1, 288–305.
46 What I mean by “one” decree here is: one level of decision in God—over
against the two in traditional Reformed orthodoxy. This is not to overlook Barth’s doc-
trine of reprobation. It is only to suggest that in Barth, the doctrine of reprobation is a
function of the doctrine of election, and thus does not introduce a distinct level of deci-
sion in God.
47 I am aware of the discussion concerning Barth’s universalism. Berkouwer’s dis-
cussion provides a good overview: G. C. Berkouwer, The Triumph of Grace in the The-
ology of Karl Barth (Grand Rapids: Eerdmans, 1956), chapter X.
 INSEPARABLE BOND 279

namely the being of God as being God in Christ.48 The event of the
preaching of the gospel and the human response to it is a mere recogni-
tion—both on the part of the preacher and on the part of the believer—
of the one single act of God’s being in Christ. There is no additional
soteriological level in which the restoration of the divine-human rela-
tionship between God and the believer is taken into account. Put in trin-
itarian terms: there is no separate level of the Spirit in the economy of
salvation.49 While motivated by a concern to build the relationship
between God and human beings into the very being of God, Barth ended
with a static account of this relationship, a relationship in which a reci-
procal action between God and the believer cannot truly be taken into
account.50

48 On this point, see especially the essays on time and eternity: CD I/2, 45–121,
and III/1, 42–93; KD I/2, 50–133, and III/1, 44–103.
49 In a sense, Barth’s critique of Cocceius as having no room for the Spirit in the
pactum salutis is a typical case of the pot calling the cattle black! Cf. van Asselt, Fed-
eral Theology, 233–236.
50 I would like to thank, in chronological order, Prof. Dr. Christoph Schwöbel, the
members of Prof. Schwöbel’s Doktorandenkolloquium at Tübingen, Dr. Bert Loonstra,
Prof. Dr. Gijsbert van den Brink, Prof. Dr. Richard A. Muller, and Prof. Dr. Marcel
Sarot for their comments on an earlier version of this paper. The research for this arti-
cle was funded by the Netherlands Organisation for Scientific Research (NWO), the
Alexander von Humboldt Foundation, and the Flemish Organisation for Scientific
Research (FWO-V).