Exegetical Fallacies by D.A. Carson

Word-Study Fallacies

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 warnings.

1. The root fallacy: The presupposition that every word has a meaningful root. Examples: αποστολος(apostle) has root αποστελλω, but this does not mean that an “apostle” is one who is sent.

2. Semantic anachronism: The mistranslation of words which have changed in meaning since the time they have been written. Example: “dynamite” comes from δυναμις, meaning power or miracle, but passages containing δυναμις can not be translated as explosive!

3. Semantic obsolescence: Instead of words being given modern meanings (as above), they are interpreted with obsolete meanings. Example: μαρτυς came to mean martyr probably after Revelations was written, having progressed in meaning from

a) one who gives evidence

b) one who gives solemn witness or affirmation (e.g. of one's faith)

c) one who witnesses to personal faith even in threat of death

d) one who witnesses to personal faith by acceptance of death

e) one who dies for a cause (modern day “martyr”)

4. Appeal to unknown or unlikely meanings: Poor or invalid research may result in the adoption of an incorrect meaning. Example: Does κεϕαλη in the singular mean “source” and “origin”, or “headship”? In all historical documents, it has been shown that “headship” is the better translation, with connotations of “authority”, not to a “river's mouth” as in classical Greek.

5. Careless appeal to background material: Similar to the previous, this does not necessarily have to involve an unlikely meaning. Examples: What sort of water is υδατος και in John 3:5? Is it amniotic fluid, a sacrament, rain, semen or simple purification? The accepted view is simply the Holy Spirit cleansing our heart and spirit while imparting God's nature to us, not the less likely semen, which has a tentative base in classical Greek. In the Sermon on the Mount, there is mention of a mountain in Matthew 5:1 but a plain in Luke 6:17. Instead of explaining it away as ‘a mountain has level places’ and being vague, it can be explained as ‘into the hill country’, capturing the meaning of πεδινος (a plateau in mountainous regions) in Luke 6:17.

*6. Verbal parallelomania: The listing of verbal parallels in some body of literature as if those bare phenomena demonstrate conceptual links or even dependency.

7. Linkage of language and mentality: There is no such thing as ‘the Greek mind’ or ‘the Hebrew mind’, and suppositions of their inability to comprehend aspects due to linguistic barriers is incorrect. Example: Just because ancient Hebrew had only verb tenses referring to temporal duration (that is, action instantaneous or action continuous), it does not mean that they had no concept of past, present or future.

8. False assumptions about technical meaning: The assumption that a certain word has a technical meaning or requirement. Example: “Sanctify” can have several meanings, whether progressive purification or an instantaneous setting aside. It cannot be given a single technical action.

9. Problems surrounding synonyms and componential analysis: Whether two words are equivalent or synonymous can only be decided by context and common sense. The difficulty is increased by the different connotations associated with a word and its ‘synonym’, and problems arise when two words related in meaning are treated as exactly the same. Examples: Describing someone as having characteristics of a man is different from someone like a man. Αγαπαω (I ‘love Christianly’) and ϕιλεω (I ‘love brotherly’) may seem to have different significant meanings, but in John 21:15-17 Jesus uses αγαπαω twice and ϕιλεω last in conversation with Peter, who only uses ϕιλεω. It is not substantiated to draw significant conclusions from the uses of these words, because in the fourth century B.C. and onwards ϕιλεωbegan to be used as ‘I kiss’ rather than ‘I love’, and so αγαπαω began to take over the meaning of it.

10. Selective and prejudicial use of evidence: This particular fallacy is one which some Roman Catholics fall into, because it is easy to think that evidence taken from all over the Bible (sometimes out of context) constitutes proof.

11. Unwarranted semantic disjunctions and restrictions: Just because A is true does not mean B is untrue, or just because a certain translation of a word fits does not mean all other possible translations are impossible.

12. Unwarranted restriction of the semantic field: Some words have many functions such as idioms which may not be known. One of the major problems of the English language is the ambiguity of the word ‘is’. It has five main uses:

a) Identity: “Is the law sin?” (Romans 7:7)

b) Attribute: “No one is good except God alone” (Mark 10:18)

c) Cause: “To be carnally minded is death” (Romans 8:6)

d) Resemblance: “The tongue is a fire” (James 3:6)

e) Fulfilment: “This is what was spoken by the prophet” (Acts 2:16)

Which usage is meant in a certain passage is sometimes hard, if not impossible to determine.

13. Unwarranted adoption of an expanded semantic field: Example: The use of church in Acts 7:38 should not be construed to mean the universal church.

14. Problems relating to the Semitic background of the Greek New Testament: Using the Hebrew equivalent of a Greek word to make a point must be justified to be valid.

15. Unwarranted neglect of distinguishing peculiarities of a corpus: This is the false assumption that one New Testament writer's predominant usage of any word is roughly that of all other New Testament writers. Examples: “Call” and “justify” are used differently by different authors.

16. Unwarranted linking of sense and reference: Not every word is a name – some refer purely to an attribute.

The Heart of the Matter: Coping with Context

Perhaps the principal reason why word studies constitute a particularly rich source for exegetical fallacies is that many preachers and Bible teachers know Greek only well enough to use concordances, or perhaps a little more. There is little feel for Greek as a language; and so there is the temptation to display what has been learned in study, which as often as not is a great deal of lexical information without the restraining influence of context. The solution, of course, is to learn more Greek, not less, and to gain at least a rudimentary knowledge of linguistics.

Grammatical Fallacies

Fallacies Connected with various Tenses and Moods

It is not altogether clear that “tense” is a very accurate way of referring to the “Greek tenses”. The problem is that some Greek tenses do have primarily temporal force (e.g. the future), and other tenses have substantial temporal force at least in the indicative mood (e.g. the aorist).

1. The aorist tense: There are two main fallacies that are caused by the misunderstanding of the aorist. The aorist is named such because it is a‑orist, meaning without a place, undefined. Its use does not only mean “once for all” or “completed”, but refers to the action without specifying whether the action is unique, repeated, ingressive, instantaneous, past, or accomplished. However, in context and relation it is sometimes possible to decide which it refers to. By assuming that the aorist only implies instantaneous and completed action, or by assuming that it is merely “punctiliar” and ignoring signs in the context which allow further analysis, the meaning of a passage can be misinterpreted.

2. The first person aorist subjunctive: There may be considered three categories of deliberative subjunctive. The true deliberative subjunctive (according to definition) is an (inter-)personal question which is asked to clarify what the problem is, for instance “What shall I do?” (Luke 20:13). The other two are pseudo-deliberative, because they ask for a response for a question which has already been partially answered in the asker's mind. “Should we pay or not pay” (Mark 12:14), asked the Pharisees and Herodians to Jesus, to force him into answering. “Shall we sin because we are not under law …” (Romans 6:15), asked Paul rhetorically. Notice that these two examples are different in that the former asks for a response from others, while the latter is answered by Paul himself. There is no fallacy in any of these categories, the points being that there is much grammatical territory that is still to be realised and defined, the results can be exegetically useful, and that some grammatical categories cover more than what they strictly mean.

3. The most common fallacy in connection with the middle voice is that it is either reflexive or suggests that the subject acts by itself. Some authors have for instance argued in 1 Corinthians 13:8 that παυσονται means that something in tongues will cause it to stop by itself (the conclusion being that there is no valid gift of tongues because they interpret to teleion as the canon). The argument is invalid by itself because παυσονται in the middle voice is always ambiguous in meaning – see Luke 8:24 where the waters επαυσαντο, not because of themselves but because Jesus told them to subside.

*4. The periphrastic perfect in Matthew 16:19: The fallacy is the assumption that all problems regarding the meaning of tenses can or must be resolved by appeal to morphological (how it is inflected or what its root is) and syntagmatic (how a word is used in relation to other words). To shed some light on difficult meanings of tenses it is also worthwhile to hold a paradigmatic examination of why one tense was used instead of another, or what differences the alternative tense would make. Many verbs in Greek are defective and lack certain tenses, which force the author to use a certain tense which was substituted at the time. In Matthew 16:19, the problem is whether (in this case) the periphrastic future perfect participles (estai dedemenon/estai lelumenon) should be translated as future perfect (will have been bound/loosed) or perfect (have been bound/loosed).

Fallacies Connected with Various Syntactical Units

*1. Conditionals: There are three main fallacies which fall under conditionals. Firstly, there is the common fallacy that if the argument is valid, the protasis (the thing assumed by the “if”) is true. The word “if” should not be substituted by “since” when ei de (first class conditional) is used in Greek. In Matthew 12:27 when Jesus asks, “Even if I cast out demons by Beelzebub, by whom do your sons cast them out?”, it does not mean that Jesus uses Beelzebub in reality. Secondly, when εαν + the subjunctive (third class conditional) is used, it does not mean that it will be fulfilled (just because the subjunctive is used). Thirdly, there is actually some significance in the tense of the verb in the apodosis (the resultant of the “if”, basically everything after the “then”), if it is taken relative to the condition's fulfilment. By using GRAMCORD (a computerised grammar concordance for the Bible) it has been shown that in the apodosis, a present indicative verb indicates action coincident with the protasis' fulfilment and future indicative or ου μη + the subjunctive implies action subsequent. check

2. The article (preliminary considerations): It is a fallacy to think that an article or lack of one in Greek means that there must be one in English, because English has two articles: definite (“the”) and indefinite (“a”), while Greek only has the definite. The following table is a bare bones guide to the correct translation of the Greek inclusion or lack of article.

3. The article (the Granville Sharp rule): Sharp's rule states that “if two substantives are connected by kai and both have the article, they refer to different persons or things…; if the first has an article and the second does not, the second refers to the same person or thing as the first… Of course the rule could also be applied to a series of three or more.” The fallacy is in taking this rule too absolutely, which has led some scholars into claiming that Matthew did not have a complete idea of the times he was writing about, since he wrote on many occasions “the Pharisees and Sadducees” instead of “the Pharisees and the Sadducees” (strictly by Sharp's rule, what Matthew would be saying that the Pharisees were not a separate entity from the Sadducees which is historically incorrect). Similarly in Revelation 2:26 the Greek is “και ο νικων και ο τηρων” (literally “and the conqueror and the obeyer”), but in actual fact it does not refer to two people but to one with both attributes. Therefore the Granville Sharp rule should not be used in its oversimplified form.

*4. The article (the Colwell rule and related matters): The Colwell rule (1933) states that if a definite noun (as reckoned by E.C. Colwell) precedes a copulative verb (γι[γ]νομαι or ειμι), it is normally (87%) anarthrous, while if it follows the verb it is probably articular. Although this rule is useful, it must be stressed that the rule works only in one direction: just because a noun is anarthrous, this does not mean that it is highly likely to be definite. This rule must be used in conjunction with other contextual indicators to have any force.

5. Relationship of tense: Exegetical and theological fallacies arise in this area when conclusions are drawn without adequate attention being paid to the relationships between two or more clauses, which are usually established by the verbal forms. For instance, in Hebrews 3:14, it would be a fallacy to state that perseverance is needed to enter God's household, because the Greek makes it clear that the perseverance is after the “having shared” (by its past tense), that the perseverance is from the result of the “sharing in Christ”.

The potential for renewed precision

With the advent of a computerised grammatical concordance, inductive analysis of New Testament Greek grammar will progress at a pace which could not have been achieved before. Grammatical constructions of any length can be retrieved from the whole New Testament painlessly, allowing greater time to be devoted to the analysis of its grammar. This will therefore contribute to the reduction of errors and to the exposure of grammatical fallacies.

Logical Fallacies

The Nature and Universality of Logic

In order to discuss Logical fallacies, it is necessary to define the term logic, for it is used in a variety of ways. The sense in which I will be using the word “logic” is the comprehensive term which refers to sets of axiomatic relationships, an evaluation of the ways of using evidence to derive correct conclusions. The other senses of “logic” are not in discussion, such as “workable”, “reasonable”, hypothetical logical arguments, “Western vs. Eastern” logic, “environmental”, the logic of political systems, etc.

A select list of logical fallacies

1. False disjunctions – an improper appeal to the law of the excluded middle: A false disjunction is where a statement is incorrectly presupposed to have only two possibilities, and that if it is not one, then it must be the other. The fallacy is that such an argument excludes a complementary of both possibilities.

2. Failure to recognise distinctions: This fallacy argues that since x and y are alike in certain respects, they are alike in all respects.

3. Appeal to selective evidence: As in word study fallacies, it is improper to exclude evidence and include others which promote the argument. Many differences between Christians can be solved if evidence is not filtered in such a fashion, presenting a one-sided view.

4. Improperly handled syllogisms: There are many syllogisms which seem at first to be valid, but on a moment's reflection they are exposed as incorrect an worthless. For example:

1. No false teaching possesses an “inherent” authority for the church.

2. Some teaching is false teaching.

3. Therefore no teaching possesses “inherent” authority for the church.

The argument is invalid because the conclusion claims to be universal, while the premises state only some teaching is false.

1. Teaching occasioned by a local situation is not universally applicable.

2. The teaching in question [in 1 Tim. 2:11–15] is occasioned by a local situation.

3. Therefore the teaching in question is not universally applicable.

The problem in this argument is not the conclusion, but the premise, for it is clearly not true.

1. Connectives such as γαρ connect their immediate context to the preceding context.

2. John 3:16 opens with a γαρ.

3. Therefore John 3:16 is connected with the preceding context.

Although this argument is valid by itself, it does not prove that John 3:16 are Jesus' words even though this verse is immediately after Jesus' recognised speech. First, the context must be shown to be Jesus' speech only, and not added by John as a conclusion based on it.

1. Present indicative finite verbs always have durative force.

2. The verbs in 1 John 3:6,9 are present indicative finite verbs.

3. Therefore the verbs in 1 John 3:6,9 have durative force.

Although the logic is correct in this example, all one needs to do in this case is to provide one counterexample to the premise, one present indicative verb which does not have durative force.

1. There are many examples where the present indicative finite verb does not have durative force.

2. The verbs in 1 John 3:6,9 are present indicative finite verbs.

3. Therefore the verbs in 1 John 3:6,9 do not have durative force.

Clearly, the premise is not universal. Thus the conclusion is wrong because it is definite, while the premise is not.

5. World-view confusion: Mental baggage, one's preconceived ideas, and current situations all affect one's judgment. For instance, to take up our “cross” to many may mean to tolerate our sicknesses or shortage of money. However, to the first-century reader such a statement would be interpreted as literal, to die in a painful, ignominious, humiliating death, which is therefore the correct interpretation. The solution to this problem is to find out more about the times in which a book was written, and interpret it the way they would have, while ensuring our own mental baggage is as far away as possible.

6. Fallacies of question-framing: “When did you stop beating your wife?” is (hopefully) an irrelevant question to any husband, so it is impossible to answer it except with a grin. “Was Reconstruction shamefully harsh or surprisingly lenient?” To actually answer this question would mean excluding the possibility that the Reconstruction may have been not particularly anything, making it a false disjunction. If all these questions seem familiar, it is because there are so many improperly framed questions (like dare I say HSC essay questions) – such traps are easy to fall into, but with care can be avoided.

7. Unwarranted confusion of truth and precision: The question, “What's the difference between accuracy and precision” is under discussion here. Precision is a measure of how little error there is, while accuracy, or truth, is when there is no error at all – not error as in mistake, but error as in “amount of vagueness”. Just because a statement is not particularly precise, this does not mean it is false.

8. Purely emotive appeals: An emotional appeal based on truth reflects sincerity and conviction; an emotional appeal used as a substitute for the truth is worthless. Extrapolations based on a single statement may yield completely different conclusions if emotions are the sole discriminator, because emotions are based on our own bias, our mental baggage. The trick is to realise the difference between an emotive appeal in addition to the truth, from an emotive appeal in order to establish a truth.

9. Unwarranted generalisation and overspecification: The fallacy in this case is in thinking that a particular can be extended to a generalisation just because it suits what we want the text to say, or in thinking that a text says more than it says.

10. Negative inferences: If a proposition is true, the negative of that proposition is not necessarily true. Even if the negative is true, it cannot be argued to be because of the positive proposition. For instance:

1. All orthodox Jews believe in Moses.

2. Mr. Smith is not an orthodox Jew.

3. Therefore Mr. Smith does not believe in Moses.

This is obviously not always true. However, consider the following:

1. All who have faith in Jesus are saved.

2. Mr. Jones does not have faith in Jesus.

3. Therefore Mr. Jones is not saved.

Although the conclusion is true, it does not follow from the argument. If the premise was exclusive (“Only all who...”), then the conclusion would be correct.

11. Unwarranted associative jumps: This particular subset of the fifth fallacy in this chapter occurs when a word or phrase triggers off an associated idea, concept , or experience that bears no close relation to the text at hand, yet is used to interpret the text. For example, in Philippians 4:13, “I can do everything through him who gives me strength” (NIV), the ‘everything’ does not include jumping over the moon or other impossible things, nor does it mean ‘everything which God wants us to do’. The latter is an unwarranted associative jump, because it does not take in the context of the passage, namely living with or without many physical necessities.

12. False statements: Sometimes, even a careful scholar can sometimes make mistakes, from incorrectly recalling the original Greek text to incorrectly stating a secondary source. In many cases these are honest errors, so it is always necessary to check up on any claims put forward.

13. The non sequitur: This refers to conclusions which “do not follow” from the evidence and the arguments presented. There are many forms, often easily presented by the syllogisms I have already constructed in this chapters, but there are many examples that are not easily analysed. One of the worst examples I have seen come from the documents of the World Council of Churches, undoubtably because many such documents are written by committee and established by consensus.

“Since ordination is essentially a setting apart with prayer for the gift of the Holy Spirit, the authority of the ordained ministry is not to be understood as the possession of the ordained person but as a gift for the continuing edification of the body in and for which the minister has been ordained.” (Baptism, Eucharist and Ministry, 22.)

[My translation: “The authority of ordained ministry” is a gift from the Holy Spirit, thus it is not a possession of the ordained person. Therefore this gift is used for continued guidance in the minister's job.]

In this claim, there are two problems. There is a false disjunction concerning “the authority of ordained ministry”, where there is an unjustified disjunction between the authority being either “a gift for service” or a “possession”. As for the conclusion concerning the “edification of the body”, this bears no obvious link to the Holy Spirit's gift. The confusing way in which these fallacies are hidden does not, in any way, elevate an untruth to valid argument.

14. Cavalier dismissal: The fallacy in this instance lies in thinking that an opponent's argument has actually been handled when in fact it has merely been written off. It may happen that an opposing opinion does not conform to a writer's own ideas or ideals, however such a difference is not a licence to dismiss the opposing opinion. Instead, such a conflict should be stated and investigated to find its source.

15. Fallacies cased on equivocal argumentation: A correct argument is not necessarily conclusive, and it is a fallacy to attribute more weight than is warranted on such equivocal arguments. An example of this is the use of a rhetorical question as an appeal to common sense may be valid but is not conclusive. A more sneaky and less commendable type of argument is one that earnestly seeks out the most ambiguous language possible in order to secure the widest possible agreement, thus nullifying any attempt to distinguish differences in opinion. Another type of equivocal argument is where more than one option is left open in the conclusion, whether intentionally or unintentionally, while making it impossible for the reader to find out which the author is saying.

16. Inadequate analogies: The fallacy in this case lies in supposing that a particular analogy is relevant when in fact that analogy is demonstrably inadequate or inappropriate.

17. Abuse of “obviously” and similar expressions: ‘It is clear why this a fallacy’. If this explanation had ended with this first statement you just read, then it is just like the reverse of a cavalier dismissal. Inadequate evidence coupled with “Nothing could be more clearer” in no way verifies the argument, in the same way a purely emotional appeal is worthless in demonstrating an opinion.

18. Simplistic appeals to authority: One should judge an exegetical work not on the basis of who said it, but whether their reasons are sufficient. Denomination-specific apologetics (e.g. Roman Catholic, Anglican or Church of England theology) cannot be used to validify a universal argument.

Keeping a lookout for logical fallacies

Like the other chapters, this one is more negative than positive; but if it results in interpreters who are marginally more self-critical in their handling of Scripture, and in readers who are somewhat more discerning when they devour commentaries, expositions, and other studies. this critique will be amply awarded.

Presuppositional and Historical Fallacies

The subject of this chapter could easily be turned into a very long book. Those interested primarily in historical fallacies cannot do better than to read David Hackett Fischer's Historians' Fallacies: Toward a Logic of Historical Thought (New York, Harper and Row, 1970), while those interested in presuppositional fallacies will have to devour substantial quantities of epistemology before they come up with satisfying comprehensiveness. Nevertheless, in a book such as this, something should be said about presuppositional and historical fallacies.

The Influence of the New Hermeneutic

Until a few decades ago, hermeneutics was largely understood to be the art or science of interpretation. The interpreter is the subject, the text is the object, and the aim in this view is for the subject to develop techniques and “feel” to enable him or her to interpret the object aright. Recently, however, the idea of the new hermeneutic breaks down this strong subject/object disjunction. This new hermeneutic argues that the interpreter who approaches the text always brings along a certain amount of cultural, linguistic, and ethical baggage – even concerning the questions asked or left unasked. This in turn becomes part of the baggage so the next time around, other questions will be asked. However, this does not mean that there can not be objective and real meaning in a text – the hermeneutic circle of changing question and response need not be vicious. The interpreter can get closer and closer to the meaning of the text as the writer originally intended it, until he or she has grasped it truly (even if not exhaustively). To do this, we must distance himself from our own background, and be prepared to bend our own opinions, values, and mental structures to what the Bible and Jesus teach. Among Christians, there should be little excuse for ignoring or avoiding what the Bible has to say, on the false ground that knowledge of objective truth is impossible.

Some Presuppositional and Historical Fallacies

1. Uncontrolled historical reconstruction: A speculative reconstruction of first-century Jewish and Christian history is what it is: an educated guess at the development of the early church. New Testament scholars have traced a network of theological trajectories to explain how the church changed its thinking from decade to decade and from place to place. The error committed by some, however, is when they apply their scheme of historical development back to the original documents. Because there is very little evidence for the development of the church, methodologically it is very dangerous to squeeze New Testament documents into a speculative time frame. Any critical judgments based on this must include for the possibility that the reconstruction is wrong, in an admission of relative uncertainty.

2. Fallacies of causation: Fallacies of causation are faulty explanations of the causes of events. Fischer lists quite a few, including post hoc, propter hoc (the mistaken idea that if event B happened after event A, it happened because of event A); cum hoc, propter hoc (mistaking correlation for the cause); pro hoc, propter hoc (putting the effect before the cause); the reductive fallacy (which reduces complexity to simplicity, or diversity to uniformity in causal explanations); the fallacy of reason as cause (which mistakes a causal for a logical order, or vice versa), the fallacy of responsibility as cause (confusing a problem of ethics with a problem of agency in a way which falsifies both.

3. Fallacies of motivation: This may be considered as a subset of causal fallacies, with the effect being the intelligent act and the thought behind it being the cause. The worst cases of this are an attempt to psychoanalyse participants in a past event, sometimes with only fragmentary evidence. Other cases include attempts to base historical developments on specific choices and preferences.

4. Conceptual parallelomania: Conceptual parallelomania is particularly inviting to those who have taken advanced training in specialised field, but who have no more than a serious Sunday-school knowledge of the Scriptures.

5. Fallacies arising from omission of distanciation in the interpretative process: The most obvious of these is reading one's personal theology into the text. Also, attempts to maintain neutrality are near impossible, and it is better to discern what our prejudices are and make allowances for them, while learning as much historical theology as possible.