In identifying the Coptic Text as a forgery, I will present two distinct lines of argument. There will be some interference between the two lines.
The first line is a close comparison of the words and more importantly the phrases of the Text with the words and the phrases of the Coptic Gospel of Thomas. Every single one of the handful of phrases (and almost every single word) in the Text is found almost exactly in the Gospel of Thomas.
The second line is a discussion of grammatical inconsistencies. The special focus will be on two grammatical blunders. A mistake is something that one makes as a student or in other transparent manifestations of human fallibility. A blunder is something done inadvertently by someone who wants to avoid detection by devious dissimulation. Sahidic Coptic is by now so well known that one can confidently identify certain ways in which one writes as fully grammatical and idiomatic Coptic and other ways in which one writes as absolutely not. In that regard, the two blunders cannot be condoned. The fact that the Text is otherwise written more or less in fairly standard Sahidic Coptic (and, I would say, borrowed from the Gospel of Thomas) only makes any stark deviations that have every appearance of being errors stand out all the more.
First Avenue: A Comparison with the Coptic Gospel of Thomas
It is altogether justified to state that every single phrase of the Text can be found in the Gospel of Thomas. Full details follow below. There are slight deviations. But importantly, these deviations do not produce what might be called a different phrase.
No one can possibly doubt that, if someone wanted to create a composite text by using only snippets from the Gospel of Thomas, something very much like the Text would absolutely have to be the exact result.
To the extent that in certain cases not enough words are found to make a phrase because the text is fragmentary or presented as fragmentary, the words also all occur in the Gospel of Thomas, with one critical exception, namely t = a-hime (my wife) in line →4.
In regard to this exception, it is interesting to note that the Text deviates from the Gospel of Thomas because the latter uses the form shime (wife). Accordingly, the expected equivalent of “my wife” in the Gospel of Thomas would be t = a-shime. In other words, in the sole word or expression, as opposed to phrase, for which the Text does not have an equivalent in the Gospel of Thomas, the Text also deviates from the Gospel of Thomas.
Before entering into a more detailed comparison of the Text with the Gospel of Thomas, it will be useful to consider issues of probability.
The great Laplace (1749–1827) already pointed out in the philosophical essay that accompanies the third edition of his epochal Traité analytique du calcul des probabilités (1820) as an introduction that critical issues of probability are so often disregarded in the humanities.10 One consideration is the devastating effect of compound probability. Compound probability stipulates that the probabilities of events need to be multiplied to obtain the probability of their joint occurrence.
Consider the throwing of three dice. The probability of obtaining a six is 1 in 6 for each of the three throws as events taken separately. But the probability of obtaining three sixes in the joint occurrence of all three events viewed as a single compound event is 1/6 × 1/6 × 1/6 or 1/216, or less than half a percent. Therefore, if three conditions need to be true at the same time for an event that depends on them to occur and the probability of each condition's existence is estimated generously at 79%, then the event has less than a 50% chance of happening as a result of the conditions on which it depends for its existence because 0.79 × 0.79 × 0.79 = 0.493039, which corresponds to about 49.3%.
One might object and say that 49% or so is still significant. For example, if there were a 49% chance that it would rain tomorrow at an outdoor event, I would be very wise to take an umbrella. If one would win $100 if a coin landed heads, an event whose probability is 50%, who would not take the bet? The propitiousness of 49% or 50%, however, rests entirely on the fact that there is no penalty for flipping tails, nor for bringing an umbrella when it ends up not raining. If it were stipulated in addition that one would have to pay $100 if the coin landed tails, one might think twice about entering said bet, and even more than twice if the chance of the favorable event were less than 50%.
Something similar applies in the quest for truth. Stating something that has a 49% chance of being true at the same time means stating something that has a 51% chance of being false. There is nothing wrong with freely contemplating all kinds of possibilities, even outlandish ones, before drawing conclusions. But why would anyone want to defend a conclusion that is more likely to be false than true? Scientists are not impressed by a hypothesis that has a chance of more than 5% of being false and often demand something close to 1% even to pay attention.
As the eyes wander from one line of the Text to the next, the reader again and again encounters phrases found in the Gospel of Thomas. And as will be evidenced below, just about every phrase—with one telltale exception—can be found in the Gospel of Thomas. It is true that there are a few slight deviations mostly restricted to a single word. A phrase and its variant with only minimal deviation still share a list of several identical features and combinatory syntactic characteristics that suffice to make them essentially the same phrase. Each word and each combination between two words and each sequential relation between two words is a separate feature. Accordingly, the number of features quickly adds up so as to guarantee total distinctiveness. The deviations do not in the least affect the striking individuality of the phrases as distinct from all other possible phrases of the Coptic language.
The obvious question naturally arises: Can this be a coincidence? Could the author have composed the text and proceeded line by line, again and again producing phrases found in the Gospel of Thomas? Naturally, if it is not a pure coincidence, then the conclusion must be that there is a definite cause or factor that triggers the recurrence of the phrases from the Gospel of Thomas. And what else could this cause or factor possibly be other than that the author is adopting phrases from the Gospel of Thomas? The Gospel of Thomas is an integral literary work that is internally fully coherent. It is therefore not possible that the Gospel of Thomas borrowed phrases from the Text.
Is it possible to quantify exactly the degree of probability that the recurrence of phrases from the Gospel of Thomas either is a coincidence or is not a coincidence? The exact probability cannot be measured. There are just so many possible phrases in the Coptic language, and no one could possibly count them or, if they are at all countable, know how many there are.
However, it is altogether possible to compute the probability of absence or presence of coincidence numerically in relation to a certain exact limit above or below which a certain probability must be located. How so?
It is a fact that there are at least six phrases in the Text that are also found in the Gospel of Thomas. There is no need to consider more as long as any other phrases in the text are also from the Gospel of Thomas. And in fact they are. The six can therefore for all practical purposes represent the entire text. Considering more than six phrases would only strengthen the final result, as will become obvious below. Then again, taking no more than six is a measure designed to keep the argument safer. It is also a way of taking no more than one needs to make a point.
Clearly, readers are witnessing the appearance of a phrase from the Gospel of Thomas six times in a row. What are the chances that this is a coincidence?
At this point, it will be useful to take a look at a case in which the sixfold occurrence of an event is surely pure coincidence. Consider someone throwing a six with one die six times in a row. If the die is not compromised, the sequence must be pure coincidence. It would be difficult for anyone witnessing such an event not to be a little baffled. One would rightly wonder and be curious: What were my chances of being a witness to such an event?
Everyone agrees that the chances can be computed exactly in the case of throwing dice. One obviously has a chance of 1 in 6 of obtaining six at each throw of a die. Therefore, the chance of throwing six exactly six times in a row is no doubt 1/6 × 1/6 × 1/6 × 1/6 × 1/6 × 1/6, or (1/6)6, that is, one-sixth to the sixth power. That is, one has a chance of 1 in 46,656 of witnessing such an event. In other words, the odds are 46,655 to 1. There is no one who doubts that, on average, one can expect throwing six exactly six times in a row only once in 46,656 sets of six throws. That is a little over 0.002%.
Now back to the sequence of six phrases from the Gospel of Thomas. In the case of the throwing of a die, everyone knows that obtaining a certain number is pure coincidence. However, in the case of obtaining sentences from the Gospel of Thomas again and again, it is not really known beforehand whether that sequence is pure coincidence or not. It is therefore necessary to evaluate two scenarios: absence of pure coincidence and presence of pure coincidence.
Let us first evaluate the presence of pure coincidence. In the case of the throwing of a die, an event involving pure coincidence, the chance of obtaining six in one throw is 1 in 6. It is the ratio of the desired outcome, namely 1, to all the possible outcomes, namely 6. But what is the chance of obtaining a phrase from the Gospel of Thomas, assuming the presence of pure coincidence? Clearly, the answer is the ratio of all the different phrases in the Gospel of Thomas to all the possible phrases in Coptic. What is this ratio?
It is not possible to know this ratio exactly. But it is easy to accept that there are at least six times as many different phrases found in Sahidic Coptic as there are phrases in the Gospel of Thomas. It is therefore possible to use the numbers obtained from the case of throwing a die described above as a higher limit of the probability. In other words, as the mind of the author of the Text selected phrases for the composition in question, he or she did something that happens only once in at least 45,656, perhaps more, cases.
However, it appears to me that seasoned readers of Coptic will readily agree that there are many times more than six times as many possible phrases in total in the Coptic language as there are phrases in the Gospel of Thomas, even if the Gospel of Thomas is of a decent length. I personally find it easy to imagine that there are at least 100 times as many possible distinct Coptic phrases as those found in the Gospel of Thomas. There are so many ways in which one can combine words with one another in distinct phrases. Likewise, combinations of just the twelve notes of the chromatic scale keep yielding ever new songs. And there are many more than twelve words in Coptic.
If we assume there are exactly 100 times as many possible Coptic phrases as there are attested phrases in the Gospel of Thomas, what are the chances that a mind that randomly selects phrases from the Coptic language obtains six from the Gospel of Thomas in a row? The answer is (1/100)6, or 1 over 100 to the sixth power, or 1 chance in 1,000,000,000,000, that is, one in a trillion, effectively zero.
In other words, in seeing phrases from the Gospel of Thomas appear six times in a row, the observer has every right to claim having witnessed an event that can take place only one in a trillion if not more times—if it is assumed that the selection is purely random.
To be clear, I emphasize again that computing the probability that the phrases of the Text are not derived from the Gospel of Thomas to anywhere near precision is impossible. What does appear possible is giving numerical expression to the fact that the probability is staggeringly low, even if it cannot be computed exactly how very low.
There are far too many variables to make sophisticated computations possible. However, it is possible to compute the probability of a simple artificial scenario whose probability must be higher than any scenario that is anywhere close to reality. And it appears that assuming artificial conditions that are as favorable as possible to a high probability that there is no connection with the Gospel of Thomas already produces an exceedingly low probability.11
One artificial scenario presented above yielded a chance of one in a trillion. In fact, again, if one assumes that the Coptic corpus has only six times as many phrases as the Gospel of Thomas does, which is far too few, the probability of encountering phrases from the Gospel of Thomas six times in a row is already at least like the probability of throwing a six with one die six times in a row, namely 0.002%. I am excluding variation in the word order of the phrases and all other real complications for the sake of the argument. It is not possible to compute by how much including them would further lower the probability. But lower the probability it clearly would.
My conclusion is that it is out of the question that the sequence of six phrases from the Gospel of Thomas found in the Text is the result of pure coincidence or random selection. Accordingly, some definite cause or factor must account for the selection of phrases from the Gospel of Thomas. And what else could account for the selection of phrases from only the Gospel of Thomas other than that someone deliberately selected phrases from only the Gospel of Thomas?
There is a possibility of lowering the probability under the assumption of pure random selection even further. But I refrain from exploring it fully. Suffice it to note the following. It is a fact that a random selection of phrases cannot be 100% random. Once a certain phrase is selected, not any phrase can come next. There has to be continuity in purport. It is impossible to quantify the matter. But it is possible that there are, relatively speaking, fewer phrases that provide continuity in purport in a limited text such as the Gospel of Thomas than there are in all of the Coptic language. The ratio between the two may well be smaller than 1/100. And so would the corresponding probability, perhaps as little as one in a quadrillion or less.
The reader of this report is still owed the positive verification that just about everything in the Text is found in the Gospel of Thomas and, I am personally convinced, was taken from it. References in what follows are to the sections of the Gospel of Thomas. The following comparison is meant to be sufficient, not necessarily exhaustive, although most of what is relevant is probably included.
Line →1: [ ] y an (not[?]). This is the end of a sentence. Suffice it to note that there is more than one sentence ending in , including na = y an, in the Gospel of Thomas. See Sections (55) and (101).
Line →1: t = a-ma’w a = s-ti na = y p-ō[nh] (As for my [t = a] mother [ma’w], she [= s] has [a =] given [ti] me [na = y] life [p-ōnh]). Compare Section (101): t = a-ma’w de m-me a = s-ti na = y m-p-ōnh (As for my true mother, then, she has given me life).
Comment: The absence of de m-me ([my] true [m-me] [mother], then [de]) in the Text does not make this statement a different phrase as the term is understood here. On the absence of before in the Text, which is grammatically unacceptable, see further below.
Line →2: [. . . peje m-mathētēs n-I(ēsou)s je s[. . . (The disciples said to Jesus: “. . .”). Compare Sections (18) and (20): peje m-mathētēs n-I(ēsou)s je. Similar examples could be adduced.
Line →3: arna. This is the last word of a sentence. The word (renounce) occurs in Sections (81) and (110), in (81) at the end of a sentence.
Line →3: Mariam mpša mmo = s a[n?] (Mary [Mariam] is not [an] worthy [mpša] of it/her[?] [mmo = s]). Compare Sections (56) and (111): p-kosmos mpša mmo = f an (The world [p-kosmos] is not [an] worthy [mpša] of him [mmo = f]).
Comment: The change from masculine mmo = f to feminine mmo = s does not make this a different phrase.
The referent of feminine s in is not clear. Did the forger misunderstand the syntax of the expression or was he or she somehow bent on changing masculines into feminines as elsewhere in the Text? I am willing to venture a provisional hypothesis regarding what I think happened. Evidently, p-kosmos (the world) has been replaced by Mariam (Mary). The meaning of the original sentence is “The world is not worthy of him.” Perhaps the forger intended to make the statement “(One) is not worthy of Mary,” replacing mmo = f (of him) by mmo = s (of her). But by also replacing “the world” by “Mary,” the forger ended up stating, “Mary is not worthy of her(self).” That would be a third grammatical blunder (for grammar, see below).
Line →4: peje I(ēsou)s na = w (Jesus said to them). See Section (14): .
Line →4: t = a-hime (my wife). This expression is exceptionally not in the Gospel of Thomas, even if the word for “wife” is present, but in the form shime and not hime (see remark above). Two more letters following t = a-hime at the end of line →4 are of uncertain interpretation.
Comment: There is no doubt that shime is by far statistically the normal form in literary Sahidic Coptic. Also, shime is otherwise used in the Gospel of Thomas. The conclusion is obvious. If 1) just about every word and phrase in the Text was taken from the Gospel of Thomas as I am convinced it was, and 2) at the same time the phrase t = a-hime (my wife) was not taken from the Gospel of Thomas, then the expression t = a-hime (my wife) must have been added to what is otherwise entirely a patchwork of phrases from the Gospel of Thomas. If one considers the forger's presumed motives, it is easy to see why this phrase specifically is the only one that is definitely an addition.
Line →5: s-na-š-r mathētēs na = y awō (she [s] will [na] be able [š] to be [r] a disciple [mathētēs] for [na =] me [= y] and [awō]). Compare Sections (55) and (101): f-na-š-r mathētēs na = y an awō (he [f] will [na] not [an] be able [š] to be [r] a disciple [mathētēs] for [na =] me [= y]), though one finds an nay instead of nay an in Section (55).
Comment: The Text has feminine “she” (s) instead of the Gospel of Thomas's “he” (f), and the Text exhibits a negated statement whereas the Gospel of Thomas has an affirmative statement. The conclusion is obvious. If just about every word and phrase in the Text was taken from the Gospel of Thomas as I am convinced it was, then the masculine “he (will be able)” must have been changed into the feminine and the negation must have been removed.
Line →6: mare rōme (people [rōme] generally do not [mare] do [this or that]). Compare Section (47): mare rōme. One letter preceding mare rōme at the beginning of line →6 is of uncertain interpretation.
Comment: The element mare is here the auxiliary verb, generally called the conjugation base (Polotsky), of the negated aorist. The typical form in standard literary Christian Sahidic is mere. But the reason for the appearance of in the Text is more than obvious. It is simply copied from the Gospel of Thomas, in which the form mare is used.
Line →6: et-how (evil [people]; literally, which [et-] is evil [how]). This expression involves both 1) an unacceptable grammatical blunder and 2) Coptic that is hardly idiomatic (more on this below). The expression does occur only once in the Gospel of Thomas, namely in Section (45), as part of the phrase pe = f-eho et-how (his [pe = f] evil [et-how] treasure [eho]).
Comment: If the author of the Text took et-how from the Gospel of Thomas, then he or she must have taken it from Section (45). As a matter of fact, I am convinced that the author did exactly that. In the Text, is followed in line →6 by šafene, which at first sight is of uncertain interpretation. However, in the Gospel of Thomas, the affirmative aorist verb form ša = f-eine (he [= f] generally [ša =] brings [eine]) occurs no fewer than three times in the very same Section (45). There can be no doubt, I believe, that the form ša = f-eine is the origin of the form šafene. But the omission of the letter i makes the form unrecognizable. What clinches the whole matter is the following. The expression et-how is preceded by rōme in line →6 of the Text and their combination signifies “evil people.” Now, in Section (45), one also finds as part of an expression signifying “evil people,” namely ou-kak[os] r-rōme. Clearly, in the grammatical debris that is line →6, we see any semblance of adequate competence in Coptic disintegrating. The expression in line →6 is somehow a single verb form that begins as a negated aorist and ends as an affirmative aorist and literally means “an evil man does not he brings.” That is a grammatical monstrosity. One is almost tempted to think: the rascal!
This incompetent blundering makes it very improbable that the forger is ancient. A modern forger could presumably easily produce the Text with only a very poor knowledge of Coptic because all sorts of tools are available such as transcriptions and translations, including interlinear translations. Accordingly, the blundering described above can easily be explained in the case of a modern forger. Whenever the modern forger leaves the confines of his or her poor grasp of Coptic, it would be natural to commit blunders. And that is what I am convinced the forger did.
The matter differs for an ancient forger. An ancient forger would have had none of the aforementioned tools. Accordingly, the ability to cull phrases knowledgeably, without any help, from the Gospel of Thomas must imply an adequate knowledge of Coptic. In fact, I know of no cases in antiquity in which Coptic scribes were not native speakers of Coptic. In other words, I know of no cases in which adequate knowledge of Coptic on the part of a scribe of literary texts is not native knowledge. Someone with an adequate knowledge of Coptic—let alone a native speaker—could not have engaged in the grammatical blundering described above. Since the author of the Text did, he or she cannot be ancient.
The fact that the Text is a patchwork of phrases taken from the Gospel of Thomas by itself already exposes it as a forgery. But grammatical errors add something more. In conjunction with what has been inferred above in terms of competence in Coptic by the patchwork character, they show that the Text cannot be ancient. And being modern by itself in its own right additionally confirms that the author is a forger.
Line →6: šafene.
Comment: For the explanation of this expression, see the preceding section.
Line →7: anok ti-šoop nmma = s (as for me [anok], I [ti] am [šoop] with [nmma =] her [= s]). Compare Section (30): anok ti-šoop nmma = f (as for me [anok], I [ti] am [šoop] with [nmma =] him [= f]).
Comment: The Text has feminine “her” (s) instead of the Gospel of Thomas's “him” (f). The conclusion is obvious. If just about every word and phrase in the Text was taken from the Gospel of Thomas as I am convinced it was, then the masculine must have been changed into the feminine.
Line →8: ouhikōn (an image). This expression is found in Section (22); the word hikōn also occurs in Section (50).