Exegetical Fallacies

Carson, D. A.

• Verlag: Baker Academic
• Jahr: 1996
• ISBN:
  9781585582808
• Auflage: 2
• Thema: Theologie & Religion
• Thema: Biblischer Bezug

This book offers updated explanations of the sins of interpretation to teach sound grammatical, lexical, cultural, theological, and historical Bible study practices. "A must for teachers, pastors, and serious Bible students."-- Journal of the Evangelical Theological Society

1

Word-Study Fallacies

What amazing things words are! They can convey information and express or elicit emotion. They are the vehicles that enable us to think. With words of command we can cause things to be accomplished; with words of adoration we praise God; and in another context the same words blaspheme him.

Words are among the preacher’s primary tools—both the words he studies and the words with which he explains his studies. Mercifully, there now exist several excellent volumes to introduce the student to the general field of lexical semantics and to warn against particular abuses;[1] and this is all to the good, for Nathan Söderblom was right when he said, “Philology is the eye of the needle through which every theological camel must enter the heaven of theology.”[2]

My own pretensions are modest. I propose merely to list and describe a collection of common fallacies that repeatedly crop up when preachers and others attempt word studies of biblical terms, and to provide some examples. The entries may serve as useful warning flags.

Common Fallacies in Semantics

1. The root fallacy

One of the most enduring of errors, the root fallacy presupposes that every word actually has a meaning bound up with its shape or its components. In this view, meaning is determined by etymology; that is, by the root or roots of a word. How many times have we been told that because the verbal cognate of ἀπόστολος (apostolos, apostle) is ἀποστέλλω (apostellō, I send), the root meaning of “apostle” is “one who is sent”? In the preface of the New King James Bible, we are told that the “literal” meaning of μονογενής (monogenēs) is “only begotten.”[3] Is that true? How often do preachers refer to the verb ἀγαπάω (agapaō, to love), contrast it with ϕιλέω (phileō, to love), and deduce that the text is saying something about a special kind of loving, for no other reason than that ἀγαπάω (agapaō) is used?

All of this is linguistic nonsense. We might have guessed as much if we were more acquainted with the etymology of English words. Anthony C. Thiselton offers by way of example our word nice, which comes from the Latin nescius, meaning “ignorant.”[4] Our “good-bye” is a contraction for Anglo-Saxon “God be with you.” Now it may be possible to trace out diachronically just how nescius generated “nice”; it is certainly easy to imagine how “God be with you” came to be contracted to “good-bye.” But I know of no one today who in saying such and such a person is “nice” believes that he or she has in some measure labeled that person ignorant because the “root meaning” or “hidden meaning” or “literal meaning” of “nice” is “ignorant.”

J. P. Louw provides a fascinating example.[5] In 1 Corinthians 4:1 Paul writes of himself, Cephas, Apollos, and other leaders in these terms: “So then, men ought to regard us as servants (ὑπηρέτας, hypēretas) of Christ and as those entrusted with the secret things of God” (NIV). More than a century ago, R. C. Trench popularized the view that ὑπηρέτης (hypēretēs) derives from the verb ἐρέσσω (eressō) “to row.”[6] The basic meaning of ὑπηρέτης (hypēretēs), then, is “rower.” Trench quite explicitly says a ὑπηρέτης (hypēretēs) “was originally the rower (from ἐρέσσώ [eressō]).” A. T. Robertson and J. B Hofmann went further and said ὑπηρέτης (hypēretēs) derives morphologically from ὑπό (hypo) and ἐρέτης (eretēs).[7] Now ἐρέσσω (eressō) means “rower” in Homer (eighth century B.C.!); and Hofmann draws the explicit connection with the morphology, concluding a ὑπηρέτης (hypēretēs) was basically an “under rower” or “assistant rower” or “subordinate rower.” Trench had not gone so far: he did not detect in ὑπο (hypo) any notion of subordination. Nevertheless Leon Morris concluded that a ὑπηρέτης (hypēretēs) was “a servant of a lowly kind”;[8] and William Barclay plunged further and designated ὑπηρέτης (hypēretēs) as “a rower on the lower bank of a trireme.”[9] Yet the fact remains that with only one possible exception—and it is merely possible, not certain[10]—ὑπηρέτης (hypēretēs) is never used for “rower” in classical literature, and it is certainly not used that way in the New Testament. The ὑπηρέτης (hypēretēs) in the New Testament is a servant, and often there is little if anything to distinguish him from a διάκονος (diakonos). As Louw remarks, to derive the meaning of ὑπηρέτης (hypēretēs) from ὑπό (hypo) and ἐρέτης (eretēs) is no more intrinsically realistic than deriving the meaning of “butterfly” from “butter” and “fly,” or the meaning of “pineapple” from “pine” and “apple.”[11] Even those of us who have never been to Hawaii recognize that pineapples are not a special kind of apple that grows on pines.

The search for hidden meanings bound up with etymologies becomes even more ludicrous when two words with entirely different meanings share the same etymology. James Barr draws attention to the pair לֶחֶם (leḥem) and מִלְחָמָה (milḥāmâ), which mean “bread” and “war” respectively:

It must be regarded as doubtful whether the influence of their common root is of any importance semantically in classical Hebrew in the normal usage of the words. And it would be utterly fanciful to connect the two as mutually suggestive or evocative, as if battles were normally for the sake of bread or bread a necessary provision for battles. Words containing similar sound sequences may of course be deliberately juxtaposed for assonance, but this is a special case and separately recognizable.[12]

Perhaps I should return for a moment to my first three examples. It is arguable that although ἀπόστολος (apostolos, apostle) is cognate with ἀποστέλλω (apostellō, I send), New Testament use of the noun does not center on the meaning the one sent but on “messenger.” Now a messenger is usually sent; but the word messenger also calls to mind the message the person carries, and suggests he represents the one who sent him. In other words, actual usage in the New Testament suggests that ἀπόστολος (apostolos) commonly bears the meaning a special representative or a special messenger rather than “someone sent out.”

The word μονογενής (monogenēs) is often thought to spring from μόνος (monos, only) plus γεννάω (gennaō, to beget); and hence its meaning is “only begotten.” Even at the etymological level, the γεν (gen)-root is tricky: μονογενής (monogenes) could as easily spring from μόνος (monos, only) plus γένος (genos, kind or race) to mean “only one of its kind,” “unique,” or the like. If we press on to consider usage, we discover that the Septuagint renders יָחִיר (yāḥîd) as “alone” or “only” (e.g., Ps. 22:20 [21:21, LXX, “my precious life” (NIV) or “my only soul”]; Ps. 25:16 [24:16, LXX, “for I am lonely and poor”]), without even a hint of “begetting.” True, in the New Testament the word often refers to the relationship of child to parent; but even here, care must be taken. In Hebrews 11:17 Isaac is said to be Abraham’s μονογενής (monogenēs)—which clearly cannot mean “only-begotten son,” since Abraham also sired Ishmael and a fresh packet of progeny by Keturah (Gen. 25:1–2). Issac is, however, Abraham’s unique son, his special and well-beloved son.[13] The long and short of the matter is that renderings such as “for God so loved the world that he gave his one and only Son” (John 3:16, NIV) are prompted by neither an inordinate love of paraphrasis, nor a perverse desire to deny some cardinal truth, but by linguistics.

In a similar vein, although it is doubtless true that the entire range of ἀγαπάω (agapaō, to love) and the entire range of φιλέω (phileō, to love) are not exactly the same, nevertheless they enjoy substantial overlap; and where they overlap, appeal to a “root meaning” in order to discern a difference is fallacious. In 2 Samuel 13 (LXX), both ἀγαπάω (agapaō, to love) and the cognate ἀγάπη (agapē, love) can refer to Amnon’s incestuous rape of his half sister Tamar (2 Sam. 13:15, LXX). When we read that Demas forsook Paul because he loved this present, evil world, there is no linguistic reason to be surprised that the verb is ἀγαπάω (agapaō, 2 Tim. 4:10). John 3:35 records that the Father loves the Son and uses the verb ἀγαπάω (agapaō); John 5:20 repeats the thought, but uses φιλέω (phileō)—without any discernible shift in meaning. The false assumptions surrounding this pair of words are ubiquitous; and so I shall return to them again. My only point here is that there is nothing intrinsic to the verb ἀγαπάω (agapaō) or the noun ἀγάπη (agapē) to prove its real meaning or hidden meaning refers to some special kind of love.

I hasten to add three caveats to this discussion. First, I am not saying that any word can mean anything. Normally we observe that any individual word has a certain limited semantic range, and the context may therefore modify or shape the meaning of a word only within certain boundaries. The total semantic range is not permanently fixed, of course; with time and novel usage, it may shift considerably. Even so, I am not suggesting that words are infinitely plastic. I am simply saying that the meaning of a word cannot be reliably determined by etymology, or that a root, once discovered, always projects a certain semantic load onto any word that incorporates that root. Linguistically, meaning is not an intrinsic possession of a word; rather, “it is a set of relations for which a verbal symbol is a sign.”[14] In one sense, of course, it is legitimate to say “this word means such and such,” where we are either providing the lexical range inductively observed or specifying the meaning of a word in a particular context; but we must not freight such talk with too much etymological baggage.

The second caveat is that the meaning of a word may reflect the meanings of its component parts. For example, the verb ἐκβάλλω (ekballō), from ἐκ (ek) and βάλλω (ballō), does in fact mean “I cast out,” “I throw out,” or “I put out.” The meaning of a word may reflect its etymology; and it must be admitted that this is more common in synthetic languages like Greek or German, with their relatively high percentages of transparent words (words that have some kind of natural relation to their meaning) than in a language like English, where words are opaque (i.e., without any natural relation to their meaning).[15] Even so, my point is that we cannot responsibly assume that etymology is related to meaning. We can only test the point by discovering the meaning of a word inductively.

Finally, I am far from suggesting that etymological study is useless. It is important, for instance, in the diachronic study of words (the study of words as they occur across long periods of time), in the attempt to specify the earliest attested meaning, in the study of cognate languages, and especially in attempts to understand the meanings of hapax legomena (words that appear only once). In the last case, although etymology is a clumsy tool for discerning meaning, the lack of comparative material means we sometimes have no other choice. That is why, as Moisés Silva points out in his excellent discussion of these matters, etymology plays a much more important role in the determination of mea...