by Dr. Vern Sheridan Poythress


Title Page
Table of Contents 

It is now time to summarize some of the basic principles involved in a symphonic approach to doing theology. I express these principles here in twelve maxims.

1. Language is not transparent to the world.

Natural human languages are not simply perfect, invisible glass windows that have no influence on what we see in the world. Nor is there a perfect language available that would be such a perfect window. In particular, no language will enable us to state facts without making any assumptions or without the statements being related to who we are as persons. No special language can free us from having to make crucial judgments on the basis of partial analogies or similarities. No special language can immediately make visible to us the ultimate structure of categories of the universe.

Positively, natural languages are adequate vehicles for human communication and for communication between God and human beings.1 Some of the features that might be supposed to be imperfections are in fact positive assets. In the Bible God uses ordinary human language rather than a technically precise jargon. He does not include all the technical, pedantic details that would interest a scholar. By doing so, he speaks clearly to ordinary people, not merely to scholars with advanced technical knowledge. What God says is not exhaustive, but it is sufficient to save us and to provide a sure guide for our life.

Hence, the ordinary, humble readers of the Bible do all right. Paradoxically, scholars and would-be scholars can easily get into trouble by overestimating the degree of technical or pedantic precision in the Bible. They will then fall into mistakes that an ordinary reader of the Bible would not make. Therefore, if we are engaged in more scholarly theological reflection, we must become self-conscious about our language. Of course, most scholars do not explicitly do theology on the basis of some fully developed philosophy of language. Mistakes made without an explicit philosophy are nevertheless mistakes. And such mistakes do occur. In fact, mistakes, obscurities, inadequacies, and infelicities related to language occur with considerable frequency in our day, even in reputable, scholarly writings of theology.2

James Barr has published a long catalog of mistakes made by biblical scholars. Without going into detail, I list six of the most common errors that he cites.3

Attempts to deduce theological conclusions directly from the grammatical structure of a language. For example, Thorlief Boman tries to deduce a philosophical concept of time from the Hebrew tense system (pp. 79-81).

Attempts to deduce theological conclusions directly from the number and relation of vocabulary synonyms. For example, Edmond Jacob attempts to deduce the fluidity of the concept of miracles from the fact that several different terms are used (p. 147).

Attempts to use etymology instead of the current meaning of a word, even when the current meaning is well known. For example, "holy" and "healthy" are etymologically related, but they do not now mean the same thing, and it is just confusing to say that they do (p. 111).

Attempts to deduce a particular world view on the basis of combining the various senses of a single word. The Hebrew word dabar sometimes means "word," sometimes "matter" or "thing," depending on the context. But Thomas F. Torrance wrongly draws the conclusion that often it means both at the same time (p. 133).

"Illegitimate totality transfer." The various meanings that a word has in all its contexts in the Bible are all read into a single passage (p. 218). For example, because the Bible teaches in various places that the church is the bride of Christ, the body of Christ, and a manifestation of the kingdom of God, people may think that ekklesia ("church") means all of these things together whenever it occurs.

"Illegitimate identity transfer." Because two words refer to the same thing, the two words must mean the same thing (pp. 217-218). For example, the Hebrew word dabar ("thing") is sometimes used to refer to a historical event, and the word "history" can also be used to refer to the same event, but it is wrong to conclude that dabar means "history." As a parallel illustration, note that we can designate the same person both as "the brightest student in the class" and as "the only redhead in my family." Although the two descriptions refer to the same person, they do not have the same meaning. "Student" does not mean the same as "redhead," or "class" the same as "family."

Barr's book and later books along the same line concentrate to a considerable extent on biblical scholars.4 But some analogous problems arise among systematic theologians, who often find themselves involved in problems related to the opacity of their own technical terms. An example or two may suffice.

Louis Berkhof, near the beginning of his discussion of dichotomy and trichotomy, says, "It is customary, especially in Christian circles, to conceive of man as consisting of two, and only two, distinct parts, namely, body and soul. This view is technically called dichotomy."5 What does Berkhof mean by a "part"? And what evidence determines whether a part is "distinct"? Arms and legs are parts of an individual, and they are distinct. Are conscience, memory, imagination, and emotions parts? Are they distinct? Berkhof's usage causes us to raise questions about the use of key terms, for instance, "Does the use of the key terms in their key contexts assumes an ontology derived from Aristotle?" Or, "Are we to think of the universe as made up of a number of diverse self-existent substances, with attributes attached to those substances, and these distinct substances later combining?" I contend here, not that Berkhof assumes Aristotelian metaphysics, but that he has not clearly identified his own assumptions.

By the very next page, Berkhof concludes, "The prevailing representation of the nature of man in Scripture is clearly dichotomistic."6 Berkhof can affirm this position only by assuming that the key terms are clear. But they are not. The result is that the whole rest of the chapter is built on air. The question at stake here is not first of all the correctness of dichotomy or trichotomy or some other view. The question is whether we even know clearly what we are saying and know what will count as evidence to support what we are saying.

Karl Barth's neoorthodox theology also exhibits violations of maxim 1. The most basic problem with Barth's whole theology, from the standpoint of language, consists in the pervasive ambiguity in the meaning of nearly every fundamental term. Barth, in the judgment of many, has made revelation a dimension of life with no direct contact with the ordinary, the secular, and the scientific world. But without this contact, theological vocabulary makes no difference in the word and threatens to become meaningless.

Thus Anthony Thiselton, appealing to Wittgenstein, observes, "Concepts like 'being redeemed,' 'being spoken to by God,' and so on, are made intelligible and teachable not on the basis of private existential experience but on the basis of a public tradition of certain patterns of behavior."7 Thiselton is criticizing Bultmann, but similar criticism might apply with less intensity to Barth. The problems can be illustrated by taking a passage almost at random from the Church Dogmatics. In the discussion of man, Barth says:

Godlessness is not, therefore, a possibility, but an ontological impossibility for man. Man is not without, but with God. This is not to say, of course, that godless men do not exist. Sin is undoubtedly committed and exists. Yet sin itself is not a possibility but an ontological impossibility for man. We are actually with Jesus, i.e., with God. This means that our being does not include but excludes sin. To be in sin, in godlessness, is a mode of being contrary to our humanity. For the man who is with Jesus--and this is man's ontological determination--is with God. If he denies God, he denies himself. He is then something which he cannot be in the Counterpart in which he is. He chooses his own impossibility. And every offense in which godlessness can express itself, e.g., unbelief and idolatry, doubt and indifference to God, is as such, both in its theoretical and practical forms, an offense with which man burdens, obscures and corrupts himself. It is an attack on the continuance of his own creatureliness.8

Undoubtedly we must try to make allowances. The larger context of this particular passage and of Barth's work as a whole can help us to have some idea what Barth is really saying. Moreover, in examining this particular passage, we must allow for the rhetorical value of stating things paradoxically, mysteriously, or hyperbolically. But my criticism is that it is too easy to agree with Barth by reading in one's own meaning or disagree with him by reading in a contrary meaning. Are the key terms such as "godless," "ontological," "possibility," "with Jesus," "with God," "include" versus "exclude," and "attack on continuance" anywhere defined?

Barth's "universalism" is visible here, a position arising from the observation that Jesus was a man and that he was "for" us. But a large amount of this universalist rhetoric simply either by-passes the issue of what the particular respects are in which Jesus is for us or else it vacillates between different senses. Hence it is easy for disputes to arise as to just what this universalism amounts to. If Jesus is "for us," does that mean that everyone is saved in the end? Or is there an "end" in the traditional sense? Or are some people (possibly) finally lost, even though the Atonement is universal? Or are there simply universal benefits to an act of atonement that also has a narrower focus on Christ's sheep (John 10)? Or is the Atonement really a kind of eternal act that is just a metaphor for God's kindness, since the Atonement seems to define creation as well as redemption?

Most of these questions perhaps show a scholastic or static orientation that is alien to the dynamic character of Barth's approach to theology. Some may object that such "static" questions are beside the point. But the problem is that if one cannot answer the static questions or justify those answers or at least say that, for such and such reasons, we do not know the answers, then the meaning is indeterminate. This dynamic approach all too easily boils down to a wax-nose theology that can mean anything or nothing. It sounds good because the words still carry emotional associations that they have acquired in a more traditional context.

2. No term in the Bible is equal to a technical term of systematic theology.

Words and phrases in the Bible are words and phrases of ordinary language. They have a meaning, but the meaning has fuzzy boundaries that are usually not as sharp as the boundaries of technical terms. And normally words used in the Bible will not, by themselves, have the rich intension that characterizes technical terms. Of course, it is natural for at least a few religious terms to develop a semitechnical precision. In first-century Judaism, the words for altar, temple, sacrifice, and Sabbath had some of this character. But it is easy to overestimate this precision. Because the vocabulary of a language must be a medium for expressing many differing and even contradictory views, it retains a generality. The views held by particular individuals or sects are much more precise than the normal meanings of the vocabulary items that they use in expressing their views.

It follows that, if we want to develop technical terms in theology, such as "Trinity" or "saving faith," we cannot make those terms perfectly match individual Hebrew or Greek vocabulary items. The reason is that we cannot make a word with a precise meaning exactly match a word with a broad, flexible meaning, or one with several shades of meaning in different contexts. If we really succeeded in making some English word "exactly match" a Hebrew or Greek word, the word in English would be just as vague and flexible as the one in Greek or Hebrew; it would not have any of the advantages of technical precision or fixity of meaning. The more precise we make the technical term, the greater distance it must have from an exact match to any one word of Hebrew or Greek.

Systematic theology can easily become the most confused in reflection on the "order of salvation" (ordo salutis). Consider the terms "regeneražtion," "vocation," "faith," "justification," "adoption," "sanctification," and "glorificažtion." All of these words, in the course of historical reflection in the church, have become technical terms designating actions of God and of human beings in the process of working out salvation. Often, in the use of these technical terms, two conflicting desires are at work: (1) a desire to match English words with vocabulary items of Hebrew or Greek and (2) a desire to condense a great deal of doctrine and historical theology into a single word.

Charles Hodge, for example, tries to satisfy both desires at once. On the first page of his section discussing the order of salvation, he describes the work of the Holy Spirit in bringing people to faith. Then he adds:

This work of the Spirit is in the Scriptures called VOCATION. It is one of the many excellences of the Reformed Theology that it retains, as far as possible, Scriptural terms for Scriptural doctrines. It is proper that this should be done. Words and thoughts are so intimately related that to change the former, is to modify, more or less seriously, the latter. And as the words of Scripture are the words of the Spirit, it is becoming and important that they should be retained.9

This passage demonstrates Hodge's lack of sufficient reflection about the functioning of language. In the first place, Hodge has mixed inspiration and translation together in a confusing way. In his doctrine of inspiration, Hodge correctly maintains that the Bible is verbally inspired, that God chose the very words of the original, and that the message composed of those words has full divine authority. This position, however, says nothing directly about a theory of translation, namely, a view of how we express the message in another language with different words and different grammar. Hodge has ignored the complexity of the process of translation and interpretation.

But Hodge is also wrong about Reformed theology and about his own vocabulary. It is easy to show that his vocabulary does not match biblical vocabulary. For example, "faith," as a technical term in the order of salvation, cannot be equated with Greek pistis or pisteuo. The Greek word pistis is used with the sense "faithfulness" (Rom. 3:3), "solemn promise" (1 Tim. 5:12), "a special gift of faith" (1 Cor. 12:9), "body of belief" (Jude 3), as well as specifically "trust" (Rom. 3:22). The verb pisteuo is used with the sense "entrust" (Luke 16:11). When it has the sense "believe" it can be used neutrally (Acts 9:26), weakly (Luke 8:13), and negatively (2 Thess. 2:11), as well as in contexts that connect it more closely with saving faith in the narrow, technical sense (Acts 4:4). The range of use of pistis or pisteuo is far from "matching" that of the technical idea of saving faith. Hodge and theologians before him had isolated passages such as Romans 3:22 and Acts 4:4 and built the technical term "saving faith" on such passages alone. The technical term does not spring full-grown from a single word, but derives from the teaching of whole passages. The technical term actually does not perfectly match the meaning of pistis or pisteuo even in the key passages, since the passages are richer than the word alone. We learn about saving faith from the joint contribution of all the words of the passage, in their context and their relations to one another--not from analyzing a single word. A technical term in English (or any language) thus does not match any one Greek word but captures common features of a number of whole sentences or paragraphs dealing with a given subject.

Matters can become quite obscure when a theologian define a technical term means or stick to one meaning. In his book Faith and Sanctification Berkouwer does not define "sanctification."10 Most of the time he uses the word "sanctification" (Dutch heiliging) in a way similar to its normal technical use within academic theology, namely, to describe the growth of the believing person in spiritual maturity and moral rectitude. At other times, however, Berkouwer appeals to passages in the Bible that happen to use the words hagios, qadosh, and their cognates, even when the words may not be closely related to "sanctification." The result is that some portions of the book appear quite muddled.

A somewhat extended example can fully illustrate this confusion. In the midst of an extended discussion about the relation of justification and sanctification, Berkouwer writes:

Understood in this fashion, the distinction between justificažtion and sanctification would amount to assigning the one act wholly to God and the other wholly to man. Sanctification would then be described as a series of devout acts and works performed by the previously justified man. The distinction between justification and sanctification could then be traced to the subject of each act: God or man.11

Berkouwer is right to object to this distorted approach. But the reasons he gives for his position are not so adequate. He says, "It is not hard to see that the Scriptures are intolerant of this division. We are told, for example, that Christ 'was made unto us wisdom from God, and righteousness and sanctification, and redemption' (1 Cor. 1:30)."12

Unfortunately, Berkouwer has not distinguished the issue of definition from the issue of the theological affirmations to be made using the term so defined. In theory, at least, we could define the English technical term "sanctification" as the actions of human beings in the course of maturing in their lives. As long as we did not deny that God was also at work, there would be nothing wrong with such a definition. It is simply a matter of history that "sanctification" is customarily used in systematic theology to describe God's action in causing people to grow.

But Berkouwer, by simply quoting passage in which God or Christ is the agent of "sanctification," proposes to demonstrate that sanctification is not the action of human beings. He demonstrates no such thing. If, in systematic theology, "sanctification" has come to mean the action of God in causing us to grow, then no quotation of the Bible is necessary to show that sanctification is the action of God. It has this meaning by definition. If, on the other hand, we are asking what the Bible says about spiritual growth, then we have to cite passages that are about spiritual growth. 1 Cor 1:30 may or may not be such a passage. The word "sanctification" appearing in some translations does not settle the question, because we must ask whether the original text contains that technical meaning. Paul perhaps had in view "consecration," the fact that union with Christ gives us the status of priests before God. The word hagiasmos in 1 Corinthians 1:30 could mean roughly what occurrences of qds often mean in the Old Testament: separation for God's service, involving declared perfection. Such usage would say that, by virtue of union with Christ, we are perfectly holy, fit to stand in God's presence (see Heb. 10:10, 14). The term, then, would come quite close to what we normally designate "justification." If so, 1 Corinthians 1:30 is irrelevant to Berkouwer's point.

By reading on from this page in Faith and Sanctification, one can easily see that the same problem occurs repeatedly. When citing a passage about holiness, Berkouwer never asks whether or in what way the "holiness" is related to the technical term "sanctification."

The general lesson here is that a lot of such mischief has arisen from an underlying fallacy. We believe that, when God speaks, he will exercise infinite care in what he says and how he says it. But then we add, "God therefore uses words with a pedantically precise technical meaning." But such is not the nature of God's infinite care. God cares enough to speak human language masterfully, which means using to the full the vagueness and variation possible in the meaning of individual vocabulary items.

This idea is really not a new one but was recognized in a general way by the classic advocates of the orthodox doctrine of inspiration. Thus Benjamin B. Warfield says:

And there is no ground for imagining that God is unable to frame His own message in the language of the organs of His revelation without its thereby ceasing to be, because expressed in a fashion natural to these organs, therefore purely His message. One would suppose it to lie in the very nature of the case that if the Lord makes any revelation to men, He would do it in the language of men....
. . . Not only was it God the Lord who made the tongue, and who made this particular tongue with all its peculiarities, not without regard to the message He would deliver through it; but His control of it is perfect and complete, and it is as absurd to say that He cannot speak His message by it purely without that message suffering change from the peculiarities of its tone and modes of enunciation, as it would be to say that no new truth can be announced in any language because the elements of speech by the combination of which the truth in question is announced are already in existence with their fixed range of connotation [note the acknowledgement of range of meaning here].13

Similarly, A. A. Hodge says, "The thoughts and words are both alike human, and therefore subject to human limitations, but the divine superintendence and guarantee extend to the one as much as the other."14

God fully knows variations in possible use of each vocabulary item of Hebrew and Greek. He does not invent an artificial jargon of technical terms but speaks using the resources that he himself has designed, the resources of vagueness. We are less careful than he is when we import mistaken views of human language and expect God to conform to them.

The fallacy can have its effect in at least two directions. We may attempt to read artificial precision into all the occurrences of the same word. We then imagine that it must mean the same thing (namely, our technical definition) in every single occurrence. Conversely, we may suppose that, despite the obvious diversity of uses of a given word, we must search for some some deep, mysterious unity of a "core meaning" to be discovered underneath. Both of these approaches manifest lack of real care in the understanding of human language.

3. Technical terms in systematic theology can almost always be defined in more than one way. Every technical term is selective in the features it includes.

By "technical term" I refer primarily to terms such as "Trinity," "imputation," "covenant," "regeneration," and "soul," which are typically introžduced by way of definitions. But many other abstract terms used in philosophy can become technical if they are used as key terms in ways that deviate from some kind of average use in the language. Such terms characteristically become filled with meaning (i.e., become rich in intension), either by associating them with explicit definitions intended to make them precise or by using them repeatedly in key contexts and turning points in the arguments. Often the aim is to use a given technical term with precisely the same meaning every time it occurs.

Technical terms, at their best, identify "natural classes," that is, a group of things sharing interesting similarities of various kinds. Once again, let us take "faith" as an example. How do we come to using "faith" as a technical term, in the sense of "saving faith," "faith in Christ that leads to eternal salvation"? We do so by seeing a consistent pattern in Acts, in Paul, and even in the Old Testament. For example, there are quite a few interesting characteristics shared by passages in Acts and elsewhere that which, in connection with people's conversion to Christianity, mention cognitive, trustful responses to a message about Christ's work and a call to repentance. In quite a few of these passages, the word pistis or pisteuo occurs (though some passages have other words such as peitho). Hence it is useful to invent the technical term "saving faith" to designate some common features of the passages.

But the grouping together could have singled out other features, depending on our interest. For instance, we could have studied occurrences of pisteuo and related words in the context of belief in idols and false gods. We could have studied the contexts in the Bible that speak of a person's entrusting something to someone else. We then could have invented a technical term "idolatrous faith" or "entrustment" to capture common features of these passages.

Or we could choose the same passages in Acts and single out different aspects. We could choose to emphasize the difference between the responses in Acts, in the post-Pentecost context, and the responses before Pentecost. The cognitive content of people's beliefs is, in general, richer after Pentecost. And the presence of the Holy Spirit in a new way in the spreading of the gospel colors the responses subtly. On the other hand, we could choose to emphasize the continuities before and after Pentecost. The technical term "faith" in Reformed theology has typically chosen this latter course.

Moreover, we could choose to focus narrowly on the striking common cognitive features in people's responses. We could focus on assent to the truth. Or we could choose to produce a technical term that designates the whole response of trust (Latin fiducia). We could produce a term that encompasses the entire change of life-direction ("conversion" or even "conversion-initiation," including baptism).

The point is that the choices are ours. No passage forces upon us only one "right" technical term. Of course, there may be several inappropriate ways of formulating a definition. That is, there may be some plausible definitions that presuppose something untrue. Such definitions are misleading. But the rejection of misleading definitions in this sense still leaves us with more than one option. Theological texts sometimes propose a single "right" definition. Underlying such an approach is the illusion that a technical term ought to and can match an item of uninterpreted biblical vocabulary. To proceed on the basis of this illusion is to invite trouble. It simply allows us to conceal from ourselves the influences of our own interests, biases, and selectivity on the kind of technical terms we use. It allows us to introduce all kinds of unexamined assumptions into our interpretation of the Bible.

Whatever way we choose to define or use a technical term, it will not capture everything. If we define the technical term "faith" narrowly, so that it applies only to the post-Pentecost situation, it will not allow us to identify the resemblances between events in Acts and those in Luke and in the Old Testament. If we choose to define "faith" so that it includes the faith of Abraham, we will not also be able to single out those features and contexts that make people's response after Pentecost unique. Moreover, the only way in which we may capture in a new term the entire range of a Greek or Hebrew word is to make our own term imprecise, with different senses or connotations in differing contexts. We must then abandon the idea of making the word a technical term in any normal sense.

4. Boundaries are fuzzy.

The world has many natural classes (genera and species) whose boundaries are fuzzy. The groups are characterized by gradual reduction in the degree of resemblance to paradigm cases, which means that, however we may choose to define a technical term, there will be borderline cases. We will not always be sure whether a given case is an instance to which the technical term applies. "Saving faith" as a technical term certainly illustrates this uncertainty, because we cannot read another person's heart. The problem, however, arises not only from the limitations of our knowledge of the world but also from the limited precision of the language used in the definition itself. We should carefully examine the language used in defining technical terms and ask what key terms mean. How precisely do we really know what is involved? Can we think of borderline cases for the applicability of key terms in the supposed definition?

5. No category or system of categories gives us ultimate reality.

No category gives us a kind of metaphysically ultimate analysis of the world. Nothing will change the fact that we are creatures with limited knowledge and with a variety of possible perspectives. In speaking of a category or theme, I have in mind not only biblical themes such as "covenant," "revelation," "prophet," "king," and "priest" but also terminology coming from other sources, such as the normative, situational, and personal perspectives on ethics, or philosophical terminology such as "being," "infinite," "necessary," "logical," "reason," "existence," and "mind."

I claim that no single category, theme, or concept and no system of categories can furnish us with an infinitely deep analysis of the world. No category gives an analysis that is innately more penetrating than any other could be. Moreover, no category is capable of being formed that allows human beings to separate the world or any aspect of the world neatly into two parts, leaving no residue or disagreement about possible intermediate cases. No category, whether from philosophy, theology, natural science, or any other discipline, gives us the essence of a particular group of things.

There are several possible ways to verify the truth of this maxim. One way is to start with meditations about the Trinity. No categories, including categories such as "person" and "substance," enable us as creatures to dissolve the mystery involved in our understanding of the Trinity. We believe that God is one and that God is three persons, but we cannot exhaustively analyze how God can be both one and three.

Neither can we make some attribute such as righteousness or self-existence into the essence of God, if we mean by such a phrase that one attribute alone can be regarded as a source of the others, and not vice versa. Moreover, God's attributes are not "deeper" than his trinitarian nature. No one attribute is the "last thing back," from which all the others are derived. Rather, any attribute can be seen as related to any other. On a human level, we can derive any attribute from any other. We just have to expand or stretch our conception of the starting attribute, and we will find that it involves or implies the derived attribute. Hence, we might just as well start with the fullness of God's existence as one God and as a fellowship of three persons, as with some attribute.

It follows, then, that no category suffices to unravel the mystery of God himself. But now what about God's creation? What God has created is not self-existent but dependent on him continuously (Acts 17:28). In a Christian or biblical world view, God himself is the only truly ultimate explanation for anything. If one category could serve as a tool for obtaining an ultimate analysis, it could only be the "category" of God. And as we have seen, God always remains mysterious for us.

Such arguments may seem excessively abstract. We can go on to consider different sorts of argument that lead to confirmation of the same principle. One such way begins with the finality and ultimacy of the revelation of God in Jesus Christ. There is no more ultimate revelation of God. Hence there is no more ultimate revelation of our significance as human beings and of the significance of our life and our world than the revelation offered of Jesus Christ in the Bible. Maxim 5 is simply a prohibition against any endeavor to go "behind" or "above" or "beyond" the Bible by means of categories that are presumably deeper or more ultimate and thus dictate the limitations of the Bible's revelation.

There are simple instances of such attempted limitations and distortions. For instance, some may classify the Bible as "religious" language and may assert that, as such, it makes no meaningful claims about events in space and time but only about some "noumenal" realm. Such claims make sense only if words such as "religious" and "noumenal," used in this way, offer some profound insight into the world. Such key terms are important aspects of a world view. This world view is in fact different from that of the Bible itself. Within the Bible there are many statements and claims about God's doing things and even predicting things in space and time. Analysis of the categories "religious" and "noumenal" within a biblical world view easily shows that there are false assumptions presupposed in such arguments.

But, more generally, philosophy and non-Christian religions are often, if not always, manifestations of full-blown world views. In these world views certain categories, which appear as key words or key concepts, form an ultimate framework for interpreting the world. The purification and advancement of Christian understanding require that we view these categories with caution. Philosophies and religions involve deep commitments, and the prior commitments influence the construction of systematic thought and key categories. Hence we need to be wary of the categories used and of how they may presuppose things inimical to the Christian world view.15 In the history of theology, the systems of Aristotle and Kant, along with variations among their followers, have been by far the most influential. Many modern thinkers have virtually begun with the observation "Of course, we are all Kantians" and have supposed that it is really impossible for anyone who has digested Kant's insights to go back to a pre-Kantian stage. The Christian, however, should critically evaluate Kant's categories right from the start. The way in which the phenomenal/noumenal distinction is construed, the way in which the category "experience" is used, and the way in which human reasoning is conceived already presuppose that the God of the Bible does not exist. Instead there is only a Kantian surrogate god who cannot appear in history, who cannot become incarnate, who cannot raise the dead, and who is alleged not to be in intimate fellowship with us in the midst of all human reasoning.

My point concerns not any one particular philosophy but the claims frequently cropping up in philosophy of many kinds. Philosophies often claim that their categories provide wisdom because they get to the roots of the world or of the human mind. I contend that nothing gets more to the roots than what God says to us. God's final word is Jesus Christ. And Jesus Christ is not to be a Christ after our own imaginations but the Christ described by the Bible.

Still another way to support maxim 5 is by appeal to perspectives. Philosophical categories and technical categories of systematic theology are often valued for their precision. But by using the technique of developing perspecžtives, we realize that sometimes this precision is an illusion and other times it is a precision achieved by drawing a boundary at an arbitrary point (we could have expanded or contracted the category). Once we see this limitation, the category loses its claim to give us some unique "depth" in our view of the world.

In particular, any category that a philosopher or theologian proposes can be converted into a perspective. We can extend it or contract it by using analogies. For instance, we may ask ourselves what Kant's category "experience," present even on the first page of the Critique of Pure Reason, really amounts to. Is it "raw" experience, before it is "processed" by our rational powers? Is it like an idealized kind of pure sensation? Or does it include everything? Does it include the experience that human beings have had of God's speaking person to person, as he spoke to Abraham and Moses? Does it include the experience that angels or demons have or that God himself has? "Experience" could embrace everything (as a perspective) or nothing (as an idealization of something prior to "processing").

Finally, we can support maxim 5 by means of a linguistic approach. In this approach, we analyze the way in which we use words in natural languages. The words have fuzzy boundaries, as do the categories we develop. We have already seen the way in which this limitation produces obstacles to the idea that we have some infinitely deep knowledge by means of infinitely sharp distinctions. Our knowledge is genuine and true, but it never makes us gods.

6. Different human writers of the Bible bring differing perspectives to bear on a given doctrine or event.

Differences between the human writers of the Bible do exist. For instance, the four gospels, written by four different human writers, show different emphases in their description of the life of Christ. Many of these differences are subtle. Others are more striking. For example, the difference between the the "I am" sayings in John and the more concealed claims in the other three gospels is noteworthy.

Even in the treatment of the same theme there may be differences. All four gospels show how Jesus fulfills God's purposes. There is harmony here. But there are also differences in emphasis and approach. Matthew is the most fond of proof-texting. He takes most care to show Jesus' relation to David, the Davidic promises, and the idea of kingship in the Old Testament. Mark relies many times on implicit similarities between the Old Testament and the events that he narrates. John focuses on the typological and symbolic relations between the work of Christ and the institutions and feasts of the Old Testament. Luke emphasizes the necessity of Jesus' going to Jerusalem and the cross, usually without citing any one Old Testament passage. The emphasis therefore falls more on fulfilling God's plan as a whole.

Suppose, then, that we tried to develop a theology of fulfillment. A definition of fulfillment or an essay on fulfillment could attempt to capture the emphasis of one gospel alone or it could attempt to weave together various features from all four gospels. In the latter case it would leave in the background the distinctive emphasis of any one gospel. No matter which way we chose to organize the discussion, we could not simultaneously do everything. There is no one "right" definition of fulfillment, nor is there any one right way of writing the essay. After all, Matthew and John both wrote essays, in a sense, and they were both right.

7. The differences between biblical writings by different human authors are also divine differences.

God uses a multiplicity of perspectives in communicating to us. We may thus view the differences between the emphases in the four gospels as divinely ordained. Hence we do not need to postulate some underlying single harmonistic account as more appropriate. Harmonization is possible in principle, but it needs to be balanced by an appreciation of divinely ordained diversity.

8. Any motif of the Bible can be used as the single organizing motif.

For the benefit of our own understanding, it is helpful to use biblical motifs explicitly as perspectives on the whole of the Bible. As an example, consider the biblical motif of justice. For the sake of gaining insight, we can treat "justice" as the most basic organizing motif of any given passage, of a book, or of the whole Bible. But in so doing, we have to be prepared to alter and enrich our definition of the motif. The theology of liberation has in fact expanded the theme of justice. But much theology of liberation, influenced by Marxist reductionism, has a distorted or truncated view of justice (a critique I cannot take space here to develop fully). It has too often contracted rather than expanded its idea of justice. In its interaction with the Bible, it has not let the Bible engage in a radical critique of Marxist ideology. The solution is not to deny that justice can be used as an organizing motif but to do the job in a way more faithful to the Bible than the theology of liberation has done.

Of course, many passages give natural prominence to one motif and not another. The second coming of Christ is pictured in one passage with a focus on judgment (Matt. 25:31-46), in another passage with the motif of war (Rev. 19:11-21), and in another passage with the motif of cosmic participation (Rom. 8:22-25). Such natural prominences should not be overlooked but taken seriously.

On the other hand, motifs are always related to one another. There is one God and one world created by God. We can therefore expect harmony in principle between different motifs. We expect that we can learn something by looking at God and the world from a variety of starting points. Consider, for example the depiction of the Second Coming in Revelation 19:11-21. We may start with the motif of holy war. This passage describes the Second Coming as involving a war: the final, climactic battle in the war between God and the hosts of Satan. This perspective is an obvious one to adopt for looking at the passage, because, as we noted above, war is a prominent motif within the passage itself.

But now suppose that we use as our starting point the motif of legal judgment. God as the Judge pronounces judgment and gives rewards or punishments to people. This process is pictured directly in Revelation 20:11-15. Revelation 19:11-21 may be describing one aspect of the same event as Revelation 20:11-15 (as amillennialists think), or it may be a separate event (as premillennialists think). But it definitely shows God's judgment implicitly. Even though the main imagery is that of war, we must remember that it is the righteous God, the Judge, who wages war, in the person of Christ (Rev. 19:11). There is implicit in the war the pronouncement of judgment on the opponents (cf. Dan. 7:25-27), and there is explicit the consignment to punishment (Rev. 19:20-21), parallel to Matthew 25:46.

We also may take the motif of cosmic transformation as basic. What happens when we look at Revelation 19:11-21 with this starting point, or perspective, in mind? The participation of the cosmos is visible in Revelation 19 in the invitation to the birds (v. 17) as well as the scope of the armies involved. And the outcome of the war, we know, is the cleansing of the whole universe by a banishment of whatever is impure (Rev. 21:4, 8, etc.). By thus using other motifs besides the most obvious ones, we can make ourselves aware of other kinds of connections in the Bible.

9. We use different motifs not to relativize truth but to gain truth.

When we follow maxim 8 above, we do not aim at relativizing truth, which would be tantamount to making truth disappear. Rather, we aim to gain more truth and to set truths in more and more complex relationships to other truths. There is harmony and not contradiction in truth.

Are the accounts of the Second Coming in Matthew 25:31-46, Romans 8:22-25, and Revelation 19:11-21 in contradiction to one another? No. There is no contradiction, even though each passage uses analogies with a different aspect of our experience (human experience of courts, longings, and wars respectively). Neither does there need to be any contradiction or relativization just because we use an analogy suggested by some other part of the Bible.

Such examples help to clarify the distinction between my approach and philosophical relativism. The goal of radical philosophical relativism is to demonstrate that there is no absolute truth, but only "truth for me." By contrast, I am producing a method of more productively proceeding toward the goal of truth. The goal is still genuine knowledge of the truth, and we can achieve such knowledge.

10. We see what our tools enable us to see.

Our ability to understand the Bible in depth depends on the gifts that God has given each one of us (1 Cor. 12:8, Eph. 4:11; 2 Tim. 2:7). In the body of Christ we are to rely on not only our own gifts but also the gifts of others. On the negative side, we may be led astray by false teachings coming from others. But on the positive side, we may profit from the genuine insights of others. The use of a variety of perspectives, in the way set forth by the previous maxims, is intended to enhance our abilities on this positive side. By deliberately using a motif that may have been noticed and developed by others, we appropriate their gifts. And we can thereby free ourselves from some of the limitations that tend to confine us when we constantly read the Bible in terms of what is most familiar to us. We may, for example, notice for the first time the note of righteous judgment in Revelation 19:11 when we try to view this passage from the perspective developed in Matthew 25:31-46.

11. Error is parasitic on the truth.

To be at all plausible, errors and lies must somehow look like the truth. They cannot sustain themselves long, and they will not be believed long, unless to some degree they disguise themselves as angels of light. (2 Cor. 11:14).

To take an extreme case, the Watchtower Society (Jehovah's Witnesses) is a heretical sect that denies the doctrine of the Trinity. But they emphasize and teach about the kingdom of God and the second coming of Christ. In this area their teaching is not the same as the Bible's, but it is close enough to have some of the attraction that accompanies biblical hope in God's promises.

Similarly, Christian Science is attractive because of its emphasis on healing. Even though it has a highly distorted view of healing, its attraction is based on a grain of truth: God is powerful even now to heal sickness when he wishes, and at the Second Coming he will put an end to all sickness and suffering among Christians.

These illustrations remind us that there is a distinction between truth and error and that some errors in doctrine are very serious. Jehovah's Witness teaching and Christian Science seriously obscure and sometimes contradict important biblical teachings that affect our eternal salvation. We ought never to forget this fact. And yet, even in such cases, we find mixtures of truth and error. It is worthwhile asking what grain of truth makes the error more plausible.

12. In theological debates, we should preempt the other person's strong points.

As we saw under the previous maxim, sometimes we are dealing with outright error, not just a harmonizable difference of viewpoint. In such cases, it is often worthwhile trying to figure out what other people fear and what are the strongest points in their arguments. We should try to find some grain of truth in their fears, in their strong points, and in the things that they care for most intensely. Even if there is only a distant similarity between what they assert and what is actually true, we can find the primary points of similarity. Starting with the actual truth closest to their viewpoint, we can develop a perspective from which to expand to the truth that we want them to learn. We can, in other words, "steal their thunder," or preempt their strong points.

Again, Jehovah's Witnesses may serve as an example. One of their strong points is their emphasis on the second coming of Christ. It might seem futile to try to outdo them at this point, for they are well trained on this subject, and have available to them their own special interpretations of a large number of texts. A discussion along conventional lines in this area is almost certain to become bogged down in endless wrangling. But we might seize on an unconventional connection. It is very important for Christians to be prepared for Christ's second coming and to understand what the Bible teaches on the subject. But how do we understand what the Bible teaches? There are false prophets who can mislead us. One of the elements in our preparation is to be on our guard against false prophets (Matt. 24:4-5, 11, 24; Mark 13:5-6; Luke 17:23). Jehovah's Witnesses are ready to agree that there are false prophets around (outside of their own group). We could then start asking about criteria for identifying false prophets (Deut. 18:18-22). Then we could go to examine the false prophecies issuing from the Watchtower Society.16

Similarly, for Christian Science one point of approach might be precisely in the area of healing. We may speak of the hope that the Bible gives us for complete healing in body, soul, and spirit at the second coming of Christ. This hope far outstrips the mental gymnastics to which "healing" reduces in Christian Science.

But situations in which we have to deal with cults are among the most difficult we ever encounter. The typical situation is one of less violent and less dogmatic disagreement. In these situations, it is even easier to show that there can be value in preempting the strong points of another.

Consider, for example, what we could say to a person who thought that evangelism was the only important thing in the Christian life. Or what about another person who thought that doctrinal purity was the only important thing? With both, we could start with the strong point. With the first, we start with evangelism itself, which is proclaiming the gospel. From Rom 1:16-17 we show how central the gospel and evangelism is but then work to show how, in Romans itself, the gospel is broader and deeper than a minimal evangelistic message. That very depth is a source of power for salvation.

For one who is oriented to doctrine, we may start with a passage such as Eph 4:11-16. Our growth in doctrine depends upon the work of each part of the body, including parts being added to the body by evangelists. Evangelism aids doctrine by bringing to the church more members who will contribute to growth. Moreover, we experience growth in the very process of rethinking the gospel to proclaim it to others. Finally, since one of the teachings of the Bible is that the church should be evangelistically making disciples, we are not doctrinally pure in a true sense unless we are actually evangelizing.


1. See Poythress, "Adequacy of Language."

2. See Carson, Exegetical Fallacies, 25-90. The main exceptions that I find are (1) exegetes who are concentrating on the meaning of a single passage and who have sound intuitions about language, (2) the great Reformation and pre-Reformation theologians, who most often knew how to look beyond the terminology to the subject matter, and (3) writers who have been trained in the tradition of analytic philosophy or twentieth-century semantics or who, by some other means, are aware of the workings of language and conceptual systems.

3. Barr, Semantics of Biblical Language, to which the page numbers in this section refer.

4. See Silva, Biblical Words and Their Meaning; and Carson, Exegetical Fallacies, which considers also problems in the writings of theologians see pp. 91-137.

5. Louis Berkhof, Systematic Theology (Grand Rapids: Eerdmans, 1939), 191.

6. Ibid., 192.

7. Thiselton, Two Horizons, 382.

8. Karl Barth, Church Dogmatics (Edinburgh: T. & T. Clark, 1936) 3. 2. 136.

9. Charles Hodge, Systematic Theology, 3 vols. (1871-73; reprint, Grand Rapids: Eerdmans, 1970), 2:639.

10. G. C. Berkouwer, Faith and Sanctification (Grand Rapids: Eerdmans, 1952).

11. Ibid., 21.

12. Ibid.

13. Benjamin B. Warfield, The Inspiration and Authority of the Bible (reprinted; Philadelphia: Presbyterian and Reformed, 1967), 93-94.

14. Archibald A. Hodge and Benjamin B. Warfield, Inspiration (reprint, Grand Rapids: Baker, 1979), 19-20. See also p. 28.

15. Cornelius Van Til has been a help to me in this area, exposing the religious roots involved in and presupposed in non-Christian philosophy. Those roots routinely involve the claim to human autonomy, and hence the rejection of the God of the Bible. By alerting us to this problem, Van Til paved the way for reevaluation and criticism of category systems that are presupposed or even explicitly invoked in theological reasoning.

16. See Robert A. Morey, How to Answer a Jehovah's Witness (Minneapolis: Bethany Fellowship, 1980).