Home » Library » Modern Library » Richard Carrier Bible

Richard Carrier Bible

Two Examples of Faulty Bible Scholarship (1999)

Richard Carrier


In response to remarks by Douglas Wilson in a debate with Ted Drange (see Wilson’s first rebuttal), I have written on two examples of how some Christians don’t understand the importance of scholarship in truly understanding the New Testament, centering around 1 Timothy. The first concerns the abuse of ancient Greek. The second concerns ignorance of the usefulness of textual criticism. These examples are relevant in themselves as typical cases which exemplify many other errors or distortions of the same kind, and how one must always be cautious of a preacher’s “scholarship,” just as one cannot trust at all in any kind of biblical “infallibility.”


Wilson wrote:

[In] 1 Timothy 2:4 God wants “all men to be saved and to come to a knowledge of the truth.” The Greek word for “all” here is “pas”, which in this context can be accurately translated either as “all men” (distributively), or as “all kinds of men.” Because that passage begins with a reference to all kinds of men (1 Tim. 2:1-2), verse four cannot be lifted from its context to mean each and every man.

This interpretation is groundless on the Greek. Indeed, the very notion that “pas” can alone mean “all kinds” in any restrictive sense like that is a complete untruth. But the passage does not even begin with a reference to “kinds” of men, but very clearly aims to include all men. Moreover, 2.4 is not grammatically related to 2.1-2. For 2.1-2 contains what Paul wants Christians to do, whereas 2.4 contains what god wants for the world. These are different contexts, and so one cannot import the context of 1-2 into 4. This is a classic baloney argument which violates all laws of grammar and logic, to get a passage to say precisely what it most definitely does not say. Yet it is a very common kind of argument, met with frequently.

The complete passage is given here in Greek with literal interlinear English. I have transcribed omega as “w” and eta as “h” and aspiration as ‘ (rather than h, so ch is c’ and th is t’ and ph is p’):

(1) parakalw oun(and so I [Paul] call upon) [you] prwton(first) pantwn(of all things) poieist’ai(to make) dehseis(petitions) proseuc’as(prayers) enteuxeis(intercessions) euc’aristias(thanksgivings) ‘yper(on behalf of) pantwn(all) ant’rwpwn(people) (2) ‘yper(on behalf of) basilewn(kings) kai(and) pantwn(all) twn(those) en(in) ‘yperoc’hi(a prominent position) ontwn(being), ‘ina(so that) hremon(a peaceful) kai(and) hsyc’ion(quiet) bion(life) diagwmen(we might spend), en(in) pashi(all) eusebeiai(piety) kai(and) semnothti(solemnity). (3) touto(this) [is] kalon(beautiful) kai(and) apodekton(acceptable) enwpion(in the presence) tou(of the) swthros(savior) ‘hmwn(our) t’eou(god) (4) ‘os(who) pantas(all) anthrwpous(people) t’elei(he wants) swt’hnai(to be saved) kai(and) to come (see last word) eis(to) epignwsin(a knowledge) alht’eias(of truth) elt’ein(to come).

There is no way this can be twisted to support Calvinism. There is absolutely no ambiguity about what Paul means: when he says all people, he means all people, and not some of all different kinds of people. Of course, he is only talking about the living, as the context establishes this to be about peace and godliness which are not relevant to the dead. The Calvinist reading is based on these three fallacies:

  • 1. Calvinists try to take the beginning of verse 2 to be a limitation on the ‘yper pantwn anthrwpwn of verse 1, but this is not what Paul states, and there is no justification or precedent in the Greek language for such a distortion. When “and” is left out of a list, this is a standard Greek rhetortical device called asyndeton, and it always means the omission of “and” and not the omission of “that is to say…” or some other logical connection, unless the logical connector being implied is used in the first connection of the series–but that is clearly not the case here. For example, Paul uses asyndeton throughout verse 1–“petitions, prayers, intercessions, thanksgivings”–and he is certainly not qualifying himself here, but asking people to do all these things, just as he asks it to be done on behalf of all people, not just kings and prominent men.
  • 2. Calvinists also try to interpret “all those in a prominent position” as referring to the elect, but this would not only require that the elect actually be “prominent”, i.e. visible (“standing out”) to all who look at them (which has odd doctrinal ramifications), but it would not make sense here. Paul is asking people to pray for peace, and he wants peace to be begged of all people, but logically it should be begged of world leaders, too, so that “kings and those in a prominent position” cannot mean anything but community leaders (he does not want to just say kings, since very few of them even exist in his time and place–he lives in an empire run by various officials, and kings only exist in border states). To say “kings and the elect” would be bizarre–why single out the elect (especially if, being elect, they will be good anyway and no one need ask it of them)? –and why mention just kings, an extremeley rare office, and not any other officials? The logic of the passage requires this to be about people in official power, who are to be petitioned for peace.
  • 3. Of course, even if we accepted all these bizarre distortions, it has nothing to do with what Paul says in verse 4. For in 1-2 he is telling people to pray for peace, on behalf of everyone alive, but in 4 he is describing what god wants, so that the use of the word “pas” (found five times in the above verses in these forms: pantwn, pashi, and pantas) in 4 is contextually unrelated to its use in 1 and 2. In 4 there is simply no other way to interpret what Paul is saying except that what god actually wants is all people, not some of all kinds of people, to be saved and to come to a knowledge of the truth.

Perhaps god wants what he actually arranged to be impossible, but that is to credit god with astonishing stupidity, and self-defeating behavior (is god neurotic?). But the plain, obvious interpretation is also supported in 2.6, where Paul says Jesus was ‘o(the one) dous(giving) ‘eauton(himself) antilytron(as a substitute-ransom) ‘yper(on behalf) pantwn(of everyone), which cannot be interpreted in any other way than that the sacrifice of Jesus, by being a ransom for everyone, is the source of salvation of everyone (who seeks salvation in Jesus, the obvious implied exception in the entire chapter). He thus did not die to atone for the sins of the elect, or for the sins of some of all kinds of people, but for the sins of all people. No other meaning is allowed by grammar or reason.


Wilson wrote:

Related to this, he points to debates about manuscript authority, and canonicity, which involve the additional issue of Church authority. When Christians affirm the inspiration and infallibility of the Bible, the question from unbelievers, “Which Bible?”, is an entirely reasonable one. The historic answer given by the Protestant faith (and I believe the Church has the authority and responsibility to give this answer) is that the Bible contains 66 books, Genesis to Revelation, and that only one manuscript tradition for each Testament is contains the canonical text. The modern plethora of translations, and competing claims of alternative readings, are not the result of the Church wielding her authority, but rather represent a usurpation of this authority by academia and big business.

This betrays an astonishing ignorance of how much he owes to scholarship. The fact is that there is no “one manuscript tradition” for any of the books of the bible, and that by reconstructing an archetype manuscript from a logical analysis of the available, often incomplete and error-filled manuscripts, scholars are actually producing a more reliable approximation to the original manuscript tradition–which is otherwise completely lost. We do not have, for example, the original letter written by Paul to Tim, but only hundreds of imperfect copies. Since all manuscripts have errors, it is not possible to pick one and say that it, and only it, is true, since it clearly cannot be. The only way to reconstruct a one, true manuscript is to engage in collation and analysis of many manuscripts. Though rarely can this be completely accomplished, it is a simple fact that a manuscript reconstructed by scholars will always be closer to what, for example, Paul actually wrote, than any existing manuscript.

As for Wilson’s implication of corporations, this can have no bearing on the Greek manuscripts themselves, which no “corporation” has anything to do with. And the fact remains that no translation, no matter how faithful, can ever truly replicate what the bible actually says in the Greek. This is a serious problem, for it means that no Christian ignorant of Greek has ever read the actual bible. Even the Muslims realize this, and hence have required that the Koran always be read by serious believers in the original Arabic. And it is not enough of a solution to merely learn Greek, for the meaning of allusions and words and grammatical constructions in 1st century Koine Greek is often inexorably tied to an understanding of how the language and associated ideas were used and understood in the 1st century. In other words, one must study Greek literature at the time, and social and economic and political history, and religious and philosophical history, to really start to grasp many of the nuances in the Greek. Wilson, for example, shows no knowledge of Greek rhetorical conventions of the 1st century in the passage analyzed above (or is deceitfully concealing such knowledge), and as I explained above, all Calvinists ignore the contextual significance of a letter being written under the Roman Empire. Proper interpretation requires such an understanding.

Now, with regard to scholars reconstructing the archetypal manuscripts of the books of the bible, I will give an example: we will use 1 Timothy again. There are several thousand Greek manuscripts of Paul’s letters, but fewer than a hundred are relevant, and among these there are a total of nineteen deviant readings in 1 Timothy alone, not counting simple spelling mistakes, which are legion. Why are only these ninety or so relevant? Because all the others are known to be copies of manuscripts we otherwise have (or copies of copies of these), and a copy is never more acurate than the text it was copied from. Thus, we need not worry about errors made in copies, since we can go straight to the “originals” (actually, since these are also copies, we call them “exemplars”). The ninety or so manuscripts that count, therefore, are all those which we know have not been copied from any other manuscripts that we have, although they are certainly copies of other copies that have been lost.

Date is often not relevant. A 15th century manuscript can still be more reliable than a 9th century manuscript, so long it is a copy from a text that we do not have: for example, the 15th century manuscript may have been made from a copy that is actually older and more reliable than the copy used in making the 9th century manuscript, or the copyists in the lineage of the 15th century copy could have been more competent and careful than those for the 9th century copy. Although an older manuscript is often better, this is not always the case, and it is folly to simply call the oldest the best (as we will see). We have no manuscript which claims to be a copy of the original, and from analysis it is known that we cannot perfectly reconstruct the original 1 Timothy from the manuscripts we do have. Of course, I should mention that there are copies in other languages, like Latin and Coptic and Armenian, which can be useful sometimes to scholars due to their great antiquity, but they are still translations, and thus cannot be directly used to reconstruct the original Greek.

The oldest copy of any of Paul’s letters comes from the Manchester papyrus, dated around 200 AD (as is standard practice, Manchester is the present location of the papyrus, not where it was found), but this is so fragmentary that it does not contain any useful segments of 1 Timothy, and neither this, nor any of several other fragmentary papyri (generally from the same century), has been of any use in resolving the uncertain readings in better-preserved manuscripts, or even in confirming the known text of 1 Timothy. The oldest complete copy of 1 Timothy is the London Sinaiticus Bible, referred to as Aleph 01, which is an “uncial” text (meaning it was written, as was the tradition before the 9th century, and as Paul would have written his original letter, in all capital letters) copied down in the 4th century AD. I will use this to study two of the nineteen deviant readings in the relevant copies of 1 Timothy to illuminate the nature of real textual criticism.

Already this, the oldest Greek version of all of Paul’s letters, shows the fallibility of scribes: in verse 4:10 Paul is usually assumed to have written that “we work hard and struggle” to be godly because we fix our hope on god. Aleph has this reading. But some other scribe crossed out “we struggle” (agwnizometha) and wrote above it “we reproach ourselves” (oneidizometha). Which is right? Did the second scribe check the copy against the original and correct a mistake? This was common practice, and most of our manuscripts have such corrections in them by a scribe’s supervisor. In the ancient world, reputable book “publishers” promissed this correction service to ensure quality, and Christians borrowed the practice to ensure accurate transmission of the word of god.

Is that what happened in 4:10? Even in 1 Timothy 2.7, a correction of “on Christ I am telling the truth” (taking the Lord’s name in vain) to simply “I am telling the truth” was made by the same supervisor, and this is known to be the more genuine reading, by comparison with other manuscripts (of course, this assumes that there was no desire to expunge blasphemy and thus alter what Paul wrote–see how tricky this is?). But this same second scribe also added letters to 3.16, converting the first word of the epigram from “he” (‘os) to “god” (t’eos), but this is believed to be wrong, by comparison with other manuscripts which were copied from older originals–but the uncertainty was so great that many of the other manuscripts also add “god” in the margin, showing that many copyists didn’t know for sure which was correct. 1 Timothy in Aleph was corrected by this supervisor a total of six times, and the original scribe even corrected himself once–and this is the oldest useful manuscript we have. You can imagine how confused things could (and did) get in later copies. Of course, all the manuscrips we know to be copies of Aleph, or copies of copies of Aleph, etc., we don’t even bother with, and the mistakes they show are not counted among the nineteen I mentioned.

As it happens, when it comes to 4:10, all the useful manuscripts are pretty evenly divided between both “struggle” and “reproach,” in no discernible way, offering no solution. Several manuscripts that are older than Aleph exist in other languages, including the famous Vulgate, as well as Syriac and Coptic editions, and they all have words which translate as “reproach” rather than “struggle,” but is this evidence of a better reading, or itself an error which became the source for the correction? We cannot know. Thus, no one today knows what 1 Timothy says here. It says one thing or the other, but that still means we cannot reconstruct Paul’s original choice. This is a common problem throughout the New Testament.

It can also be shown that Aleph, though the oldest Greek manuscript, is not always the most reliable. In 1 Timothy 1.1 Paul writes “according to the command (epitaghn) of God.” Aleph, however, says “according to the promise (epaggelian) of God.” So why do almost all bibles today ignore Aleph? Because all other relevant manuscripts say “command,” even those we know were copied from originals that were older than Aleph. Those copies in other languages, which are earlier than Aleph, also confirm the word “command.” Thus, it is the accepted reading, and by every rational argument is far more likely to be what Paul actually wrote, so anyone who treats Aleph as the one and only manuscript will not have what Paul wrote. Scholarship was indispensible to this solution. If no one had bothered to isolate the relevant manuscripts and their history, and to find out which manuscripts were copies of texts we already have, we might still not know which was the correct reading–for many copies have the reading of Aleph…because they are copies of Aleph! And manuscripts rarely tell us directly what edition they copied from. This must be figured out with historical and paleographical analysis, a complicated business.

I have examined a mere two of nineteen problems, in 1 Timothy alone, a book that takes up less than four pages of English in the New American Standard Bible. There are one thousand, four hundred and thirty eight significant deviations (again excluding spelling errors; Barbara Aland, et al, The Greek New Testament, 4th rev. ed., United Bible Societies, 1994, p. 2) in the whole of the Greek New Testament. Of those, I estimate almost a third, like the problem in 4:10, cannot be resolved with any certainty, even after the full exertion of critical scholarship and paleography. But it also follows that the vast majority of these deviations have been solved by that very same scholarly analysis. If anyone were to reject this work, and instead cling to only one manuscript, they would end up with hundreds of obvious errors, mistakes which can be proven to be erroneous by an analysis of the relationship of manuscripts to their common sources, and with the help of other techniques of paleography, such as the study of common scribal mistakes. To reject the progress made by this scholarship is to assert that only one copy of the bible, copied from a copy of copies, by one petty scribe, hundreds of years after the original, with all its obvious mistakes, is the one true bible, and to declare that what the original authors wrote is of no concern. I would ask of any Christian who wants to do this: tell me which manuscript is the one, true manuscript. I doubt any of them can even name a manuscript. But I know that none of them will choose the right one, because I know there isn’t one.

I asked Douglas Wilson to respond to what I had written above. Due to a busy schedule he could only answer in brief. His reply is reproduced here:

The Greek word pas. Thayers lists as his second definition “any and every, of every kind (A.V. often all manner of).” Liddell & Scott include among [their] meanings “of all kinds.” I believe there is room for disagreement on the grammar of 1 Tim. because pas also means every one distributively. But it is simply unfounded to represent this understanding of the translation issues as an example of “preacher scholarship.”

Re[garding] the textual criticism issues: there is a vast difference between a manuscript and a manuscript tradition. I was referring to the latter. Anyone interested in pursuing this issue simply has to interact with Ted Letis’ work on the subject. His book is called The Ecclesiastical Text.

Textual work must be done, but the modernistic spirit in which most of the work is done today is simply a hothouse for unbelief, and in most cases is driven by that unbelief. The fact that most evangelicals do not recognize this is, given the state of evangelicalism, not surprising. The search for the autographs will end just as all the searches for the historical Jesus end, in liberalism and rank unbelief. I am interested in defending the canonical text, which we have, rather than the autographic text, which we shall never have.

To the above I will append my response here:

Wilson is showing here the very ignorance of the technicalities of Greek that I want all people to be warned of when anyone makes doctrinal arguments from grammar. Ultimately, if Wilson can offer no other example in all of Greek literature of a genuinely restrictive meaning “of all kinds” for pas alone, then his argument is, as I said, “groundless on the Greek.” And so it is. For the Liddell & Scott entry (C) is clear: pas takes a distributive sense only in conjunction with numerals, and several examples are given. There are no examples in all 26 or so entries of pas in Liddell & Scott, which have the meaning “all kinds” in any other context than when paired with numerals. No numerals appear in 1 Timothy, and thus Wilson is ignoring an important rule of basic Greek grammar. In fact, if he was an experienced Classicist, he would know that this is an idiom: it takes on a peculiar meaning precisely because of its peculiar context, a peculiar context that is lacking in this passage.

Moreover, the idiom only appears “restrictive” in English translation. A Greek would not have understood it that way, and the fact that Wilson understands it that way proves my point that knowing Greek is not enough: you must know the idioms and milieu of the Greek writer’s world. We say “we admire all peoples” and this means peoples of all kinds: it includes all people, but also means “of all kinds” because, when all people are considered, they are of many kinds, and the purpose of such a statement is to express tolerance for all kinds of people, but certainly not in a restrictive sense. Thus, the sentence “we admire all peoples” is distributive (it means “of all kinds”), but not restrictive. It excludes no one, or at least is never intended to. This is exactly what “he gave all ten gifts” means in Greek: it means every conceivable kind of gift. It is distributive, but not restrictive. There is no example in all Greek literature of pas taking on a restrictive meaning by itself, or even when paired with numerals in this unique distributive idiom. It would only ever mean that when the context establishes it: for instance, if Paul had said simply pantwn alloiwn “of all kinds” or even pantoiwn which also means “of all kinds,” either of which we would expect if he had any intention to make such a vital doctrinal point. As Smyth’s Greek Grammar (section 1174.b-c) states: “In the predicate position…[or] without the article” pas means “all (conceivable)” or “every (conceivable).” This is simply what the Greek means. It never means anything else.

As for the issue of a “canonical text,” that is not entirely relevant to my remarks: for every church has its own canon. The canon is a subjective selection based on doctrine and opinion, and we already know that. There is no objective, scientific way to determine the “canonical text,” and therefore this has nothing to do with what the Bible actually says. What the Bible “actually says,” no matter which books you want to read (apocryphal or canonical, by whatever tradition), is a fairly objective, scientific question answered by palaeographical analysis, which is called “textual criticism.” It is the same methodology applied to all ancient texts–the principles are the same. There is room for opinion in this field, as with all sciences, but only when there are no facts to decide between alternative choices. And not only are the choices objectively set (and thus not decided by doctrine or opinion), every opinion-based choice made from among them is open to falsification by future discoveries which factually rule out the choice. These are all features of a science.

When we look at this, we see that the problems in the text of 1 Timothy noted above remain, as do similar problems in every other book of the Bible. We still don’t know exactly what the Bible says, and thus Wilson’s defense fails to meet the argument. Wilson claimed that the “modern plethora of translations, and competing claims of alternative readings…represent a usurpation of [church] authority by academia and big business,” and I aimed to show above that this was not true: competing claims of alternative readings are based on factual questions of palaeography which no church opinion can change, and that no big business has any hand in. The authority of “academia” that he alludes to is rooted in objective, scientific principles and facts to which Wilson can have no rational objection. He seems to assert that his church’s opinion is more important than academic research, when the truth is in fact the exact opposite: no opinion can ever compete with the facts, when there are facts to establish.

As for the “plethora of translations,” this is the inevitable result of competing churches and church opinions and ideologies (of which Wilson’s is no more authorative than any other) who insist on maintaining the notion of “biblical infallibility” while only using translations of the actual text which is purportedly infallible. The problem with this is not, as Wilson would have it, that there are many translations and only one “true” one (his, of course). The problem is that it makes no sense for a church to use any translation at all. If what the Bible precisely says is really so important, then it ought to be read in its own language, with a background in the culture, history, and literature of the time and place in which it was written. Translations, after all, are by definition interpretations and thus opinions of what the text says. No one should express any surprise at this, nor should anyone act as if any translation can prove more than what the translator thinks the original language said. And so long as people like Wilson can think the Greek says something that it clearly never did to any Greek speaker of the time, we have every reason to be suspicious of any translator with an ideological agenda. And that was the central point of this entire essay.

Ultimately, Wilson has said nothing about which “manuscript tradition” we are to believe when we have no objective way to adjudicate between them. Although originally disparaging the objective methods of adjudication, he now seems to agree that they are valuable, and with that he seems to agree that there is no one, completely correct reading known to us. Yet he also seems to stand by his previous assumption that the church is, in effect, allowed to claim that there is such a one, true reading. But the church can only do so by groundless fiat: they will decide according to what they want the bible to say, rather than what it actually says, in every case where we really don’t know what it says. But this puts any church’s choice on a par with every other. It is simply, and always will be, nothing more than opinion.

Now, I do agree that there are many polemicists who will abuse the techniques of critical scholarship for one aim or another. This is precisely what this essay is a warning against. Even atheists are not immune to this vice, and both they and theists are often victims of such abuses by others whom they cite. But this is true in any field. The textual criticism which actually gets represented in critical editions of the biblical text is not of this polemical kind, but is the result of real scientific research. Far from being “liberalism and rank unbelief,” critical editions do not omit competing information (one important sign of objectivity) but catalogue all of it, and they do not select from among alternatives according to any criteria of doctrine or religious opinion, but solely on palaeographical evidence (another important sign of objectivity). Any good translation which attempts to be as objective as possible will do the same. To go beyond this is to escape from fact, and to enter the realm of pure opinion. And no one should act as though opinions ever carry the same weight as facts, yet this is precisely what the religious among us do.

all rights reserved