The Emphatic Hypernegation That Was(n’t): Revisiting οὐ μὴ and New Testament Translation in Light of Research and Contemporary Linguistics

Abstract

The pleonastic hypernegation οὐ μὴ is widely recognized as conveying an emphatic “no.” However, all major English translations fail to render it consistently with such emphasis. This article explores the nature of this disparity by locating οὐ μὴ linguistically, semantically, and lexically within New Testament literature and contemporary research. It concludes that, despite theoretical exceptions and the erroneous trend of translations, οὐ μὴ should, in the New Testament, always be rendered with some explicit emphasis.

Keywords

double negation hypernegation linguistic negatives οὐ μὴ pleonasm negation

Introduction

Despite many advances in the study of Koine Greek and the New Testament, there remains confusion surrounding the force, use, and translation of the pleonastic hypernegation (or “double negative”) οὐ μὴ. This combination of negators, which appears about ninety-five times in the New Testament,¹ is regularly said by contemporary grammarians, linguists, and lexicographers to communicate an emphatic “no” (i.e., “no!” “certainly not,” etc.; see Table 1 for examples). Few exceptions to this rule are ever noted. However, English translations bear virtually no consistent rendering of the construction, whether within book groups in the same translation (e.g., Gospels, Pauline corpus) or for the same occurrence across a variety of translations. On average, roughly half of all NT renderings in translations are without any emphasis at all.² The observation of this disparity between the lexical data and translations is not new (see Ballantine 1897)³ nor is the reason for it apparent—though it may be due, . . . in part, to a flawed methodology and/or lack of research on this topic in general. What is clear is that lexicographers appear certain about how οὐ μὴ should be rendered, whereas translators appear certain that the lexicographers are mistaken.

Table 1.

Renderings of οὐ μὴ in Matt 10.42 (emphatic element shown in boldface)

Trans.	Emphatic?	Matthew 10.42
CEB	Yes	I assure you that everybody who . . . will certainly be rewarded.
ESV	Yes	truly, I say to you, he will by no means lose his reward.
GNB	Yes	You can be sure that whoever gives . . . will certainly receive a reward.
KJV	Yes	verily I say unto you, he shall in no wise lose his reward.
NABRE	Yes	amen, I say unto you, he will surely not lose his reward.
NASB	No	truly I say to you, he shall not lose his reward.
NET	Yes	I tell you the truth, he will never lose his reward.
NIV	Yes	truly I tell you, that person will certainly not lose their reward.
NJB	Yes	then in truth I tell you, he will most certainly not go without his reward.
NLT	Yes	you will surely be rewarded.
NRSV	No	truly I tell you, none of these will lose their reward.
REB	Yes	I tell you this: that man will assuredly not go unrewarded.

The following study addresses this disparity by relocating οὐ μὴ lexically and linguistically in conjunction with contemporary research. It argues that, despite theoretical exceptions, οὐ μὴ should, by default, be rendered with some kind of emphasis in NT literature—most likely a sense of certainty. The article begins with a brief statistical introduction to οὐ μὴ in the New Testament and a lexical survey. It then presents the (dissenting) data from translations before examining the various dimensions of linguistics to explain the logic of emphatic and nonemphatic positions.

Summary statistics

The oὐ μή construction occurs about 95 times in 86 verses⁴ in the Greek New Testament (and 660 times in 555 verses in the LXX). It occurs with either the future indicative (about ten times) or aorist subjunctive.⁵ A semantic difference between the two remains uncertain.⁶ Figure 1 displays the number of occurrences of οὐ μὴ in the New Testament and Figure 2 indicates density or concentration.

Figure 1.

Oὐ μή in the New Testament

Figure 2.

Concentration of οὐ μή in the New Testament

Several observations can be noted at this point:

Oὐ μή is heavily concentrated in the Gospels and less frequent in the Epistles, being present sparsely in only one-third of them.

Oὐ μή is exclusively discursive in the Gospels, always appearing in the speech of a character.

Following the lead of the LXX’s use of οὐ μὴ in divine pronouncements and prophetic writings, over half the occurrences in the New Testament come from the sayings of Jesus (56 out of 95; Mitiku 2005, 100).

All occurrences in the Gospels come from Jesus except the following:

Matt 16.22; 26.35; John 13.8; Mark 14.31 (Peter’s words)

Mark 16.18 (textually disputed)

John 8.52 (Pharisees citing Moses)

John 20.25 (Thomas)

54 out of 59 occurrences of οὐ μὴ in the Gospels occur in denials or predictions.⁷

Luke uses οὐ μὴ the least of the four gospel writers, including very rarely in Acts.

Some of these observations are undoubtedly owing to the fact that οὐ μὴ occurs only with the aorist subjunctive and the future indicative in the New Testament. They might also suggest that οὐ μὴ is more likely to be found in a lower register (i.e., everyday discourse) than in a higher register, like letters (i.e., more formal contexts).⁸ Emphatic expressions tend to be less common in the latter. However, οὐ μὴ is semantically complex, and one cannot simply and definitively associate a word or expression with a single register.

Other noticeable trends should be mentioned, such as the following:

The highest concentration of οὐ μὴ comes in Revelation and 1 Thessalonians, both of which are substantially eschatological.

Oὐ μή is entirely absent from 2 Corinthians, the four Prison Epistles, 2 Thessalonians, the three Pastoral Epistles, James, 2 Peter, the three Letters of John, and Jude.

The unsettling disjunction between lexicographers and translators

Whereas the late nineteenth-century work of Ballantine (1897), (early) Goodwin (1890),⁹ and Moulton (1906, 189–91) carved out space for a non-emphatic position, modern lexicographers of Koine Greek overwhelmingly see οὐ μὴ as emphatic negation. This is likely due to advances in NT research,¹⁰ to linguistic scholarship that highlights its idiomatic status as a specific word-pair, and to the lack of grammatically appropriate alternatives (cf. “Horn’s Division of Pragmatic Labor” in Horn 1984, discussed below), and is reinforced by contextual appropriateness and undisputed implementation (cf. Horrocks 2014, 45).

Whatever the reasons, the consensus is clear. BDAG calls it a “marker of reinforced negation,” having “the effect of strengthening the negation,” and “the most decisive way of negating something in the future” (735, 646). Similarly, LN calls it “a marker of emphatic negation—‘by no means, certainly not’” (1:664), Mounce, “strong negation” (2009, 295), and Swanson, “by no means, emphatic marker” (1997, s.v. οὐ). House and Robinson’s Analytical Lexicon contains, “not at all, in no way (a double neg. strengthening the denial)” (2012, 256), and the Brill Dictionary of Ancient Greek, “to express absolute negation or prohibition” (Montanari 2015, 1500). Daniel Wallace even goes as far as to say that οὐ μὴ is “the strongest way to negate something in Greek” (1998, 468). Other scholars concur that the double negation is or can be used for emphasis (Goodwin 1900, 289; Swetnam 1998, 399; Burton 1900, 174–84; cf. Thompson 1898, 371). Virtually none of the aforementioned works mentions exceptions to this conclusion. Some recent works even deny such exceptions—such as the frequent claim that οὐ μὴ has lost its emphasis due to overuse.¹¹ Additionally, exegetical commentators from a variety of approaches regularly highlight the emphatic force of οὐ μὴ.¹²

It is altogether puzzling, then, that an exhaustive tabulation of οὐ μὴ renderings in major English translations reveals a marked contradiction to this position.¹³ No translation consistently renders οὐ μὴ with emphasis. Few come even close to the ninety-five occurrences; the average is about forty-eight (see Figure 3).

Figure 3.

Frequency graph of emphatic renderings of οὐ μὴ in NT translations (out of 95 occurrences)

Furthermore, there is no pattern to this trend regarding the type of translation. The choice to render emphatically is indiscriminate with regard to formal, dynamic, or paraphrase translations. It is also difficult to discern any pattern regarding specific verses or to trace any pattern to grammatical case (e.g., subjunctive vs. future uses). There is a slight trend regarding the age of the translation; newer ones appear more sensitive to the emphatic rendering (e.g., NIV, NET, NLT, CEB) than older ones (e.g., KJV, NASB, NABRE), but it is not clear why this is the case.¹⁴

There are, however, seven noticeable occurrences where all translators chose to withhold an emphatic rendering (the following are from NRSV):

“When they persecute you in one town, flee to the next; for truly I tell you, you will not [οὐ μὴ] have gone through all the towns of Israel before the Son of Man comes.” (Matt 10.23)

“Do not judge, and you will not [οὐ μὴ] be judged.” (Luke 6.37)

They said, “If you are the Messiah, tell us.” He replied, “If I tell you, you will not [οὐ μὴ] believe.” (Luke 22.67)

“and if I question you, you will not [οὐ μὴ] answer.” (Luke 22.68)

Jesus answered, “Will you lay down your life for me? Very truly, I tell you, before [οὐ μὴ] the cock crows, you will have denied me three times.” (John 13.38)¹⁵

Jesus said to Peter, “Put your sword back into its sheath. Am I not [οὐ μὴ] to drink the cup that the Father has given me?” (John 18.11)

“Lord, who will not [οὐ μὴ] fear and glorify your name? For you alone are holy. All nations will come and worship before you, for your judgments have been revealed.” (Rev 15.4)

These results are peculiar, at least when one asks what might have driven all translators (across decades of time, geography, scholarly philosophies, publishing goals, etc.) to consciously withhold an emphatic rendering of οὐ μὴ.

There are five potential clues. Aside from the one occurrence in Revelation, of the remaining six occurrences: (a) all occur in Gospel discourse; (b) all occur in sayings of Jesus; (c) most occur in the context of some kind of foretelling/prophecy; and (d) all take place in a context where Jesus’ identity is a subject of the discourse. A potential fifth pattern (e) is clustering (οὐ μὴ occurs multiple times in a context), such as in Mark 13.2 (twice), Luke 6.37 (twice), and Luke 22.67-68 (twice). This leaves one asking: What is it about Jesus—especially his relation to the future—that would encourage translators to “water down” οὐ μὴ?¹⁶ And does the presence of clustering encourage translators to believe that the force of οὐ μὴ has lost currency due to overuse?

It is difficult to tell at this point, but preliminary answers might be constructed as follows before we turn to linguistic considerations. First, (c) may be explained by ideological motivations. It is not uncommon for translations to exhibit theological bias. In this case, such bias may amount to giving Jesus “wiggle room” regarding the definitiveness or certainty of cataclysmic future events, potentially opening up more expansive interpretations of the Olivet Discourse.¹⁷ Though unpersuasive, this remains a possibility.¹⁸ Regarding (e), it is possible that repetition would not result in more emphasis, depending on one’s linguistic framework (cf. “Joos’s Law”—more words equal less meaning—in Silva 1994, 153–54). Translators might therefore glance at two cases of οὐ μὴ in short order and conclude that there was not emphasis in either. The only problem with this explanation is that it runs directly contrary to what is found in the half-dozen occurrences of οὐ μὴ in Rev 18—which, ironically, happens to be the strongest agreement across translations for emphatic rendering of double negation (see Figure 1).

Oὐ μὴ is consistently rendered with emphasis in English translations in over a dozen passages—most from John and Revelation (Matt 24.21; John 4.14; 6.35b, 37; 8.51, 52; 10.28; 11.26; 13.8; Heb 8.12; 10.17; Rev 3.12; 18.14, 21, 22a, 22b, 23a, 23b, 25). The occurrences in the Gospels, here again, take place in Jesus’ sayings, noticeably in the context of Jesus’ identity and others’ relationship to him. As alluded to above, it is somewhat puzzling that translators agree so strongly on emphatic renderings in Rev 18 when they are so repetitive.¹⁹

In summary, at least when compared to one another, there appears to be no rhyme or reason in English translators’ choice of rendering οὐ μὴ. I have been unable to find any coherent methodology behind their decisions. One does not want to assume the worst, but it is difficult to escape the conclusion that English translators have been arbitrary when it comes to translating this construction.²⁰ Little has changed since Ballantine made the same observation over a century ago—except that scholars now firmly believe that οὐ μὴ is emphatic in the New Testament.

What remains to be examined is how οὐ μὴ is described by contemporary linguistics and what this might suggest about the dissonance between lexicographers and translators. This analysis is undertaken below, first by locating Koine Greek’s place within the context of multiple negation.

Multiple, hyper-, and pleonastic negation

Multiple negation, like negation in general, is a system that exhibits a complex constellation of linguistic phenomena with many variations and semantic functions. There are competing models and frameworks for understanding multiple negation, some more useful and complete than others. For example, in a recent article, Porter and Land (2019) apply Systemic Functional Linguistics (SFL) to ancient Greek negation. However, despite its specific application to Greek, there appear to be few or no unique implications for multiple negation beyond standard linguistic theory; a few generic categories of multiple negation are simply delineated (“negative spread,” “double negation”) with examples.

In contrast, Horn (2010, 111–12), leaning on the seminal work by Jespersen (1917), identifies a number of specific subcategories of multiple negation across multiple languages. His work also addresses motivating factors, function, and choice (all of which are central to SFL) for each category and instance. Most importantly, each of Horn’s categories of multiple negation can be readily found in Koine Greek, all of which makes his framework particularly robust for any discussion of multiple negation. For that reason, I have outlined Greek cases of multiple negation according to his primary categorization (Figure 4), which will better serve to frame our discussion of οὐ μὴ.

Figure 4.

Multiple negation

The outline below attempts to summarize and clarify this framework with examples from English and NT Greek:

Logical Double Negation, or DNA (duplex negatio affirmat) is where two negatives formally cancel each other out, which results in a positive meaning.²¹ One variant of DNA is virtual contrariety (i.e., litotes), where the force of two negatives is not entirely canceled out;²² the most typical exemplars of this subtype are not un- constructions.²³ Litotes DNA therefore overlaps with hypernegation (see [2] below) and is not consistently categorized by linguists:

DNA: She didn’t not tell me. (= She did tell me.)

DNA: ἢ μόνος ἐγὼ καὶ Βαρναβᾶς οὐκ ἔχομεν ἐξουσίαν μὴ ἐργάζεσθαι; (or is it only I and Barnabas who don’t have the right to not work for our living?)²⁴

Litotes DNA: He wasn’t unkind.²⁵

Litotes DNA: οὐ γὰρ ἄδικος ὁ θεὸς. (For God is not unjust.)²⁶

οὐκ ἀσήμου πόλεως πολίτης (a citizen of no insignificant city).²⁷

Hypernegation (DNN, duplex negatio negat) is where two negative elements combine to express a single negation. In Horn’s model, DNN comes in two major subcategories, negative concord and pleonastic negation:

Negative Concord (NC) occurs when two negatives are used in a clause but the semantics of only one negative is meant (Horrocks 2014; Labov 2001, 78, 92) and no nuance (i.e., emphasis) is present.²⁸ The negation is accomplished either sententially through a negating auxiliary (e.g., “wasn’t,” “haven’t,” “didn’t,” etc.) or modal (e.g., “mustn’t,” “wouldn’t,” “shouldn’t,” etc.), or by the use of neg-words (“no,” “not,” “neither,” “nobody”). Some of the most common examples of NC come from colloquialisms of a lower register (e.g., “I ain’t got nothin”; “I can’t get no satisfaction”), and fulfill different grammatical functions. For example, in καὶ οὐκ ἔφαγεν οὐδὲν (“he didn’t eat nothing,” Luke 4.2), οὐκ is an adverb (negating the clause) and οὐδὲν an adjective (negating the object). The most common syntactic patterns of NC are:

Neg)Aux/mod + neg-word εἰ μὴ ἦν οὗτος παρὰ θεοῦ, οὐκ ἠδύνατο ποιεῖν οὐδέν. (If this one weren’t from God, he couldn’t do nothing.)²⁹

Neg-word(subject) + (neg)aux/mod ἀλλʼ οὐ δύναται οὐδεὶς εἰς τὴν οἰκίαν τοῦ ἰσχυροῦ εἰσελθὼν τὰ σκεύη αὐτοῦ διαρπάσαι . . . (but no one ain’t able to enter a strong man’s house and plunder his property without first tying up the strong man)³⁰

Neg-word (embedded clause) and (neg)aux/mod (main clause) (Dahl 2010, 29) καὶ οὐ μέλει σοι περὶ οὐδενός (and you don’t concern yourself about no one)³¹

Neg)aux/mod in both clauses + neg-word in one of them. οὐ δύναται ὁ υἱὸς ποιεῖν ἀφʼ ἑαυτοῦ οὐδὲν ἐὰν μή τι βλέπῃ τὸν πατέρα ποιοῦντα (the Son can’t do nothing on his own, if not what he sees the Father doing).³²

Pleonastic Negation (PN) occurs when two or more negatives are used (often across clauses) but only one negative is meant. Through the use of multiple, specific negators and fewer auxiliaries and modals, the meaning is often semantically nuanced (e.g., emphasized, intensified, dampened, narrowed, “may also raise or lower the register” [Horn 2010, 132], etc.), as opposed to simple redundancy. PN is sometimes called “expletive,” “sympathetic,” “paratactic,” or “abusive” negation (Horn 2010, 121), comes in many varieties, and is used for many purposes. In general, PN “tends to reinforce or highlight the speaker’s attribution of undesirability, uncertainty, or emotive affect toward the proposition in question” (Horn 2010, 141; cf. Jackson 2016), and may give rise to idioms. It is not always easy to distinguish PN from NC (e.g., John 6.63; 15.5), nor is it easy to cleanly separate from litotes.³³

I cannot go no further (= an intensified/accentuated version of “I cannot go further”).

ἡ σὰρξ οὐκ ὠφελεῖ οὐδέν (the flesh doesn’t benefit nothing).³⁴

ὁ δὲ Ἰησοῦς οὐκέτι οὐδὲν ἀπεκρίθη (But Jesus made no further reply).³⁵

Μηκέτι εἰς τὸν αἰῶνα ἐκ σοῦ μηδεὶς καρπὸν φάγοι (No one will ever again eat your fruit).³⁶

Ὅρα μηδενὶ μηδὲν εἴπῃς (See that you say nothing to anyone).³⁷

The libertarians aren’t, I don’t think, going to support the drone-bombing bill (resumptive negation).³⁸

καὶ σκοτία ἐν αὐτῷ οὐκ ἔστιν οὐδεμία (And in him there is no darkness whatsoever).³⁹

Triplex negation confundit: There isn’t a person there who doesn’t think they can’t do it.

οὐδὲ ἁλύσει οὐκέτι οὐδεὶς (even [logical ascensive] no one could restrain him no longer).⁴⁰

Οὐ μὴ σε ἀνῶ οὐδʼ οὐ μὴ σε ἐγκαταλίπω (I will certainly not leave you nor, certainly not, forsake you).⁴¹

Another aspect of multiple negation that should be considered is its lifecycle. Like all linguistic phenomena, multiple negation is not a static phenomenon, but a dynamic and living one. The “Jespersen Cycle” is one of the most well-known ways of explaining the evolution of multiple negation—especially emphatic negation.⁴² The Cycle proposes a historical process of weakening and strengthening: “the original negative adverb is first weakened, then found insufficient and therefore strengthened, generally through some additional word, and this in turn may be felt as the negative proper and may then in the course of time be subject to the same development as the original word” (Jespersen 1917, 4). Though not all linguists incorporate the Cycle into their models, diachronic analysis and such “weakening” of multiple negation remain well-known (Givón 2001, 268; van der Wouden 1997, 245; Horrocks 2014). This weakening would seem to be the most straightforward explanation for nonemphatic uses (and translations) of οὐ μὴ in the New Testament.

Although they do not use οὐ or μή as examples, Kiparsky and Condoravdi (2006) have revised and applied the Cycle to other Greek negators (Figure 5). Their research demonstrates how “the weakening phase instantiates grammaticalization. It consists of the ‘bleaching’ of an emphatic negative into a plain negative, with loss of compositionality, and typically with phonological reduction as well” (180).⁴³

Figure 5.

Jespersen Cycle in Greek negation (Kiparsky and Condoravdi 2006, 181)

Kiparsky and Condoravdi conclude that “an emphatic negative always weakens by itself (it is ‘bleached’), never in virtue of being combined with some other element. On the other hand, negation is strengthened only by combining a simple negative with an indefinite. A simple negative, or a simple indefinite, never becomes an emphatic negative on its own” (Kiparsky and Condoravdi 2006, 193; emphasis original).

Where does οὐ μὴ fit into all of these models? First of all, there is little question from the foregoing discussion that οὐ μὴ constitutes PN.⁴⁴ It simply does not fit any other general category of DNN, including NC, because of numerous cases of undisputed emphasis.⁴⁵ It is therefore inappropriate to compare οὐ μὴ to a lower register NC like “He didn’t say nothing to nobody” (Moulton 1906, 189–91). Second, if placed within the Jespersen Cycle and current consensus about emphasis, οὐ μὴ in the New Testament would—with apparently little movement since the LXX—belong to the strengthening and not weakening phase. However, it is questionable whether this generalization is appropriate given the peculiarities of the phrase (see next section); not all double negations are equal. Whether οὐ μὴ is a clear example of the Cycle at work or not (this is an area for further study), and despite existing as a visible doublet, could it still have “weakened” by the time of the New Testament? To answer this question, one must look even deeper into the grammatical, syntactical, and idiomatic nature of οὐ μὴ.

The grammar and syntax of οὐ μὴ

As observed above, it is not uncommon to achieve emphatic negation by the use of two Greek negators (BDF, §430–432).⁴⁶ This includes NT cases of pleonastic hypernegation, like the adverb οὐκέτι used with the adjective οὐδέν (Luke 20.40; Mark 7.12), the former of which specifically denotes the temporal dimension.⁴⁷ In this case, the rendering “But Jesus made no reply any further” (Mark 15.5) would bring out the force of each negative in the sentence as a whole, while the rendering “But Jesus made no reply at all” would not be accurate. The same is generally true for οὐ μὴ, and it happens to be a pairing with a complex semantic background that virtually requires an emphatic force.⁴⁸

The semantic similarity and difference between οὐ and μή (and its various forms) has long been debated and cannot be fully recounted here. But a few key observations should be recapitulated. Though the exact origins of οὐ μὴ are unknown, there is little question that by the time of the early LXX the combination of οὐ and μή was used pleonastically—and, indeed, idiomatically—in the phrase οὐ μὴ. This contrasts with the simple use of the two (including variants of οὐ) in the same sentence (e.g., Acts 4.20; 20.20, 27). The direct sequence of one after the other is marked and has semantic import (as the aforementioned lexical works noted).⁴⁹ The overlap between οὐ and μή is substantial—enough that the force of each cannot always be distinguished.⁵⁰ But there are also subtle differences between them, and these differences strongly indicate an emphatic PN use rather than NC.

First, the two negators function in opposite ways in the context of rhetorical questions. Decker states the rule plainly: “In questions phrased in the indicative mood, οὐ implies a positive/yes answer, but using μή implies a negative/no answer” (2015, 108). He provides a comparable example in the English questions, “You studied for this exam, didn’t you?” (implying a positive answer) vs. “You didn’t study for this exam, did you?” (implying a negative answer). Although the basic binary structure of both interrogatives remains the same (and the answer to one is the answer to the other), the ultimate effect of the negative is different because of the implied answer.

Second, μὴ generally appears with non-indicative forms and οὐ with indicative mood forms (Porter 1992, 281).⁵¹ Combined with studies of negators in Greek philosophical writings (Moorhouse 1959, 37–38), it has been said that μή⁵² is the negation of the “notional or ideal” and οὐ the negation of the “concrete or actual” (Moorhouse 1959, 40). Others try to capture this distinction in different but similar terms: “μή says that one thinks a thing is not, οὐ that it is not” (LS, 507); μή is “the negative of the will and thought, as οὐ of fact and statement” (LSJ, cited in Moorhouse 1959, 40; cf. Thompson 1898, 345–46); “μή is the broader and less informative negative particle, whereas οὐ negates ideational meanings in relatively more specific environments” (Porter and Land 2019, 5); with participles, μή is used with conditions and οὐ with stating a fact (Willmott 2013, 320). It is possible that these distinctions are remnants of a stronger delineation between the two terms in earlier Greek, which have largely (though not entirely) dissolved by the time of the first century (Moorhouse 1959, 38–40; cf. Goodwin 1890). The exact relationship between the syntactical use of the two words and their semantic force remains an area for further research (cf. Willmott 2013, 328).⁵³

But the important point is that these distinctions between the negators (e.g., indicative vs. non-indicative use; actual vs. potential) make οὐ μὴ the perfect pair not only for emphatic negation, but for a certain kind of emphatic negation: the expression of certainty. Emphatic PN is used in situations where normal (or single) negation in the speaker’s idea or assertion is deemed insufficient for the communicative task. Here, the use of οὐ μὴ for the subjunctive mood and for the aorist and future tenses makes sense because they identify the cause of insufficiency. In the case of the subjunctive mood, the reality of the verbal action is conditional or potential. And in the case of the aorist and future tenses (which exhibit a “simple” or “stative” aspect), the verbal action is undefined (Hübner 2018, 107).⁵⁴ Oὐ μή is naturally summoned to duty in this unique situation to stress the certainty of the verbal action, to “seal the deal” as it were, in case there were any questions—as there might be in such a context.⁵⁵ The typical emphatic renderings “certainly not,” “never,”⁵⁶ “by no means” (and the unused, but appropriate gloss, “impossible”) in current translations are therefore appropriate.

This high degree of specificity may distinguish οὐ μὴ from ordinary cases of the Jespersen Cycle. It is generally arbitrary which two negatives are used in the Cycle, as is the word order and type of emphasis. But οὐ and μή stand as the two siblings behind the two major families of Greek negatives, function as an idiomatic pair used in a distinct order, and express a specific kind of emphasis. This specificity also highlights a main reason for the current scholarly consensus of an emphatic PN rendering: the lack of appropriate alternatives. As noted above by Kiparsky and Condoravdi (2006), ordinary negation is not adequate to express emphasis. Similarly, emphasis within NC is generally a category error of linguistics. Alternative combinations of PN hypernegation also have a different effect and emphasis than οὐ μὴ, as already illustrated in the examples of PN above; there, the PN focuses on such aspects as time (“no further,” “no more,” “no longer,” “anymore,” “ever”) and quantity (“nothing,” “no one,” “not one”):⁵⁷

A principle such as Horn’s Division of Pragmatic Labour again explains the meaning of emphatic negation: if the speaker has the choice to express a negative meaning either in a simple way (using one negation) or in a more elaborate way (using more than one negation), then the hearer may assume that the speaker doesn’t use the more complex form for nothing. Given his [or her] knowledge of language, the hearer may assume that the speaker wants to convey some sort of emphasis. This doesn’t necessarily imply that the borderline between emphatic negation and other types of multiple negations is always easy to draw. (van der Wouden 1997, 245)⁵⁸

On “weakening” and register

Is there any significant evidence that οὐ μὴ weakened—and did so by the time of the New Testament? In a word, no—largely because it is difficult to determine when and how this would ever be the case. Oὐ μή remains a marked form specifically used for the kind of nuance described above. Some have attempted to provide theoretical examples of weakening within the New Testament. “It seems probable that the force of οὐ μὴ has worn down in the LXX and the N.T.,” Robertson remarks, and cites Matt 25.9, Mark 13.2, and John 18.11 as examples of when the force is “probably weakened a bit” (1914, 1174). But how this is known is unclear and unsubstantiated,⁵⁹ and he ultimately concedes that, in general, “the force of οὐ μὴ is still strong” (1914, 1174). There have been other remarks about the changing role of each negative. Porter has argued that μή began to encroach on the territory of οὐ (1992, 281), and Willmott that οὐδὲν was also encroaching on οὐ (2013, 307). But this does not go far, as thousands of uses of οὐ were (evidently) sufficient to convey simple negation. Furthermore, despite earlier suspicions, the emphatic force of οὐ μὴ in the LXX is widely acknowledged. Recent reference works speak of it as “typical of a higher register” (Dhont 2017, 295), “somewhat forceful and made to sound solemn” (Muraoka 2009, 512), implying “strong neg. in certain sentences” (LEH), and “can be used to express a dramatic or strong prohibition” (Büchner 2017, 150).⁶⁰

What about register or diatype?⁶¹ This subject is important for almost any discussion of linguistics, but it is challenging for οὐ μὴ as it is found across registers in the New Testament—from the oral parables of Jesus, to Paul’s Letter to the Romans, to John’s Apocalypse. The same goes for the LXX, where οὐ μὴ appears over six hundred times across most books and subgenres.⁶² This does not mean it necessarily has the same effect in each register or literary context, but this does make it difficult to simply associate οὐ μὴ with a specific register.

There remains, however, an undeniable delineation between high frequency use of οὐ μὴ in Gospel discourse and low frequency in higher registers.⁶³ The most straightforward conclusion from this fact is that the frequent use of οὐ μὴ, not the idiom οὐ μὴ in and of itself, can be legitimately associated with lower registers. In other words, some expressions mean the same thing across registers, but not all registers welcome their habitual implementation. Emphatic negations like “Hell no!” and “Never again!” can be found “on the street” as well as “in the conference room,” but they are found (a) much more frequently in the former than the latter, and (b) the emphasis is felt in both cases (perhaps especially the latter).⁶⁴ If οὐ μὴ is an emphatic idiomatic negation of certainty that is common in—but not exclusive to—a lower register, this is precisely the kind of distribution one would expect across NT literature. That is, indeed, what is found. But if οὐ μὴ functions in the NT period as nothing more than a popular colloquialism, one would not expect it to be found at all from (for example) the pen of Paul or the formal pronouncements of the Apocalypse.⁶⁵

Conclusion

A recent work on Greek idioms concluded that “there has been much debate whether an aorist subjunctive or future form with οὐ μὴ is emphatically negated,” but, ultimately, “Most grammarians believe οὐ μὴ is in some way emphatically negative” (Porter 1992, 283; Porter and Land 2019, 19). In light of what has been observed in this article, this conclusion—which is the current consensus of lexical reference works and the academic community—appears to stand.

This conclusion implies a deficiency in the way English translations have rendered οὐ μὴ. Contemporary research strongly—if not conclusively—suggests that οὐ μὴ is (a) idiomatic, (b) pleonastic hypernegation, (c) emphatic, and, as I have suggested, (d) emphatic probably most often in a sense of expressing certainty. If there is any suggestion that οὐ μὴ should be rendered any differently, no coherent argument exists in this direction, and there are no criteria that have been offered to discern how and when this is the case.

Further research on this topic might answer (a) if οὐ μὴ follows the Jespersen Cycle in the transition to Byzantine Greek, and (b) if there are other similar negator pairs in the history of Attic, Koine, and Byzantine Greek that have followed the predictions of the Cycle as well. Until then, it remains valuable for students of the New Testament to always verify if οὐ μὴ is hidden beneath an ordinary English negation in their translations. Similarly, professors of Greek may be tempted to let nonemphatic οὐ μὴ translations slide on their students’ homework since “everyone is doing it.” In response to this possibility, it is hopefully clear by now what the appropriate reply should be: Oὐ μὴ.⁶⁶

Abbreviations

BDAG Bauer, Danker, Arndt, Gingrich 2000 (in References)

BDF Blass and Debrunner 1961 (in References)

CEB Common English Bible (2011)

DNA logical double negation (duplex negatio affirmat)

DNN hypernegation (duplex negatio negat)

ESV English Standard Version (2001, 2016)

GNB Good News Bible (1976, 1978 [DC], 1994 [2d])

KJV King James Version (1611)

LEH Lust, Eynikel, and Hauspie 2003 (in References)

LN Louw and Nida 1988 (in References)

LS Liddell and Scott 1996 (in References)

LSJ Liddell, Scott, and Jones 1996 (in References)

LXX Septuagint (Rahlfs–Hanhart, 2006)

NA28 Nestle–Aland Novum Testamentum Graece, 28th ed. (2012)

NABRE New American Bible, revised edition (2010)

NASB New American Standard Bible (1971, 1995)

NC negative concord

NET NET Bible (New English Translation; 2001)

NIV New International Version (2011)

NJB New Jerusalem Bible (1985)

NLT New Living Translation (1996)

NRSV New Revised Standard Version (1989)

PN pleonastic negation

REB Revised English Bible (1989)

RV Revised Version (1885)

SFL Systemic Functional Linguistics

UBS5 UBS Greek New Testament, 5th ed. (2014)

Footnotes

ORCID iD

Jamin Andreas Hübner

1

This figure excludes the construction μὴ οὐκ (found in Rom 10.18; 1 Cor 9.4, 5); the rarity of this construction makes it difficult to know if it is a semantically-equivalent alternative. All general NT references and statistics in this article are derived from the text of NA28/UBS5.

2

Emphatic renderings are explicitly marked and, fortunately, easy to identify. See .

3

Ballantine observed this inconsistency in translation in the late 1800s (mainly between KJV and RV), though he came to a conclusion opposite to that of the present article.

4

This number includes the spurious occurrence in Mark 16.18 and the occurrence in Gal 4.30 with postpositive γάρ (οὐ γὰρ μὴ).

5

The future indicative occurrences in the New Testament are Matt 15.6; 16.22; 26.35; Mark 13.31; Luke 21.33; John 4.14; 6.35b; Gal 4.30; Heb 10.17; Rev 18.14. Sometimes each mood is used in parallel without any apparent distinction, such as in John 6.35-37.

6

Campbell (2008, 59) notes from case studies that, “from the interchangeability of the future and aorist subjunctive in such constructions, it would appear that the futurity and modality of the future tense-form overlaps with the modality of the subjunctive mood, and that the perfective aspect of the future indicative parallels the perfective aspect of the aorist subjunctive.” As suggested below, this lack of clear delineation in οὐ μὴ may be due to the pair’s idiomatic status. For more on modality and negation, see .

7

See Table 3 in .

8

Some in contemporary linguistics draw a distinction between “style” and “register.” For example, , 280) state: “Styles are often analysed along a scale of formality. . . . Registers, on the other hand, when they are distinguished from styles, tend to be associated with particular groups of people or sometimes specific situations of use. Journalese, baby-talk, legalese . . . could all be considered examples of different registers. The term ‘register’ here describes the language of groups of people with common interests or jobs, or the language used in situations associated with such groups.” Others, however, see style as a parameter of a supercategory like register. For this article, I use the term “register” more generally to include both register and style.

9

William Goodwin argued that it is “impossible to defend the . . . view” that the two negatives “combine to make a single strong negative” (1890, 65–76). Ten years later, he reversed his position in his revised grammar (1900, 289), where οὐ μὴ is “emphatic” and may “express a strong prohibition” (being rendered in examples as “never” and “ever”).

10

Moulton contended, for example, that the Greek of Revelation and the Gospels is more influenced by the LXX than Paul’s writings, which—like many other observations characteristic of that era—is debatable.

11

So , 168): “we view all of the NT uses (mostly in what Jesus is reported to have said and LXX quotations) of the doubly negated aorist subjunctive as genuinely emphatic. . . . Οὐ μὴ with the aorist subjunctive has not lost its force.”

12

E.g., Barrett (2004, 238, 658); Carroll (2012, 210); , 199).

13

My sample includes CEB, ESV, GNB, KJV, NABRE, NASB, NET, NIV, NJB, NLT, NRSV, and REB.

14

Possibly against this trend, Mounce (see , 89–90) went from defining οὐ μὴ as a “very strong negative” to just a “strong negative” from the first to the second edition of his Greek grammar (1993–2003). However, I asked Mounce directly why this changed, and he replied in an email (sent April 28, 2019) that there was not a specific reason, although he added, “I think they weaken, as a general rule, the thrust of the head word.”

15

In cases like these, NRSV has inverted the negation to a positive assertion for smoother reading (from “The cock will certainly not crow until . . . ” to “Before the cock crows . . . ”). Cf. Matt 15.6; Luke 18.7. In tabulating the renderings, it appeared that GNB inverts negation more frequently than other translations. The use of the word “all” in such cases is perhaps ambiguous with regard to emphasis (e.g., Luke 21.32; Matt 24.35; etc.); I did not include these cases as emphatic in my statistics.

16

Most translations also withhold emphatic renderings for Jesus’ specific predictions in the Olivet Discourse.

17

Regarding the synoptic question as to whether the gospel writers themselves exhibited such a perspective towards Jesus’ sayings about the future, it is noteworthy that with regard to the cases above: (a) οὐ μὴ in Mark 13.2 is found in Matt 24.2, but not in Luke 21.6; (b) οὐ μὴ in John 13.38 is found in Mark 14.30 and Matt 26.34, but not in Luke 22.34. If the phrase is a lower register, it would seem Luke is “smoothing over” the discourse by removing the double negation from the earlier tradition (cf. Luke 12.11-12//Mark13.11//Matt 10.20; Luke 8.10//Mark 4.12//Matt 13.14; Luke 21.23-34//Mark 13.19//Matt 24.21; Luke 12.39-40//Matt 25.9; Luke 22.34//Mark 14.30-31; Matt 26.33-34; note, also, the transpositional textual variants in Luke 22.17-20). There are substantial exceptions, however. The double negative is retained in such places as Luke 21.33 (Mark 13.31//Matt 24.35); 12.59 (Matt 5.26); 9.27 (John 8.51-54//Mark 9.1//Matt 16.28); 13.35 (Matt 23.39); Luke 22.18 (Mark 14.25//Matt 26.29); and most notably, in 21.33 (Mark 13.31//Matt 24.35). Synoptic studies therefore shed little light. Cf. , 97–98).

18

If this were the case, one would expect the bias to emerge in certain translations (e.g., CEB, NIV) and not others like NRSV. But this is not the case. Furthermore, it does not seem like an effective apologetic to remove emphasis from a negative statement that will always remain as such. Whether a person says, “It will not rain tomorrow” or “It will certainly not rain tomorrow,” this probably does little to change that person’s credibility if it does (or does not) rain tomorrow.

19

It is perhaps significant that several in this list of emphatic renderings across translations (John 4.14; 6.35b; Heb 10.17; Rev 18.14) contain about half of the future indicative uses. Although weak, this somewhat aligns with the idea that emphasis resides primarily in the future tense uses of οὐ μὴ.

20

And, perhaps, translators have been uncritically accepting of previous renderings without asking questions about the status quo, simply for tradition’s sake.

21

Willis, Lucas, and Breitbarth (2013, 30): “This is as in propositional logic, where two negatives equal a positive: ¬ ¬ p ↔ p.”

22

, 273) also draws a potentially important distinction between intensification and emphasis, such that “emphasis always results in intensification of a kind, but that intensification could emerge without emphasis.” For simplicity’s sake, I do not bring to bear this fine distinction on my study.

23

Cf. οὐ πᾶς (in Matt 7.21); οὐ πολλοί, πολλαί (in Jas 2.12); etc.

24

1 Cor 9.6, my translation. Cf. 1 Cor 12.15. The cancelling of the negatives does not mean that the negatives can simply be removed to achieve the nuance of the meaning. In 1 Cor 9.6 (cf. Acts 4.20), making the sentence positive would require a change along the lines of, “Is it only I and Barnabas who must work for our living?”

25

“As Bosanquet (1885: 281) put it well over a century ago, ‘The essence of formal negation is to invest the contrary with the character of the contradictory,’ so that ‘From “he is not good” we may be able to infer something more than that “it is not true that he is good”’ (1885: 293)” (Horn 2010, 119). On litotes, see Horn (2015). (Note: The Bosanquet quote can be found in , 306, 310.)

26

Heb 6.10, NRSV. Cf. Jas 2.12; Acts 1.5; 17.4, 12; 19.23; 27.20; 21.39; Phil 1.17.

27

Acts 21.39, my translation. (Note NRSV and CEB’s rendering of “an important city,” which while true, misses the nuance of the litotes). Cf. Acts 26.19.

28

Cf. de Swart (2010, 34): “All in all, there is insufficient evidence that emphasis plays a role in negative concord languages in general.” Cf. Green and White-Sustaíta (2015, 435–36): “negative concord on its own—no matter how many negative elements occur in the sentence—is not a strategy for emphasizing negation . . . the use of ‘no’ does not contribute additional negative meaning; it contributes to the emphatic interpretation of the sentence [‘I don’t like no picture’].” See also “Joos’s Law” in , 153–55).

29

John 9.33, my translation.

30

Mark 3.27a, my translation. “Not” (οὐ) in this example, before a verb, functions as an auxiliary.

31

Matt 22.16, my translation.

32

John 5.19.

33

Contrast, for example, Horn’s categorization with .

34

John 6.63, my translation.

35

Mark 15.5.

36

Mark 11.14, CEB.

37

Mark 1.44, my translation. Cf. Rom 13.8.

38

In resumptive negation, “the supplementary negative is added outside the frame of the first sentence, generally as an afterthought” (Jespersen 1917, 72–73, cited in , 129).

39

1 John 1.15, my translation. Contrast the same phrase without the resumptive negation in 2.10, 3.5, and John 7.18.

40

Mark 5.3, my translation.

41

Heb 13.5, my translation.

42

In , emphatic negation can be either a minimizer (quantitative, as in “not one second” and “not a hair”) or a maximizer (qualitative, as in “not ever” and “nobody whatsoever”). However, it is not always clear if all emphatic negations fall into these categories (e.g., “not at all,” “definitely not”), especially since temporal and spatial categories are not always easily quantified.

43

They assert that “every plain negation of Greek was once an emphatic negation, at least in so far as its origin can be determined” (174). Note that Goodwin (1890) argued the opposite: “expressions with οὐ μὴ were always colloquial,” and initially “the real negative force of μή was in abeyance” (his narrative suggests an evolution from denial, to denial of future, to prohibition). Similarly, says, “however it originated and however it may have been previously used, it was not emphatic for Hellenistic writers.” All of these assertions are overstatements, in my estimation, and few (if any) contemporary scholars would agree with them (see remarks below on οὐ μὴ and the LXX).

44

Some have erroneously split the pair so that the construction is “a form of litotes, i.e., the second negative (μὴ) negates the subjunctive verb . . . ; the first negative (οὐ) negates the doubtful clause introduced by μὴ” (, 6). In evaluating the writings of Plato, Goodwin came to a similar conclusion regarding litotes, but noted that this usage is “confined to the present subjunctive” (1890, 72). This theory will be addressed more below.

45

Even Moulton conceded that οὐ μὴ had emphasis in Matt 16.22; 26.35; Mark 14.31; Luke 1.15; John 11.56; 13.8; 20.25 (, §14, 23).

46

See also μηδενὶ μηδὲν in Mark 1.44 (cf. Rom 13.8), discussed in , 119).

47

More specifically, it denotes “the extension of time up to a point but not beyond” (BDAG, 736).

48

The combination of the two negators can occur in either order, although οὐ μὴ is far more common in NT Greek than μὴ οὐ (ninety-five vs three). Mὴ οὐ also occurs in 1 Kgdms 20.3 (LXX; i.e., 1 Sam 20.3). But as indicated earlier, this strict grammatical separation is disputed, as the words may work together as an idiom. (Also note Porter’s comment below on the increasingly expansive use of μή.)

49

Efforts to split them up into different functions have generally failed. For example, Goodwin (1890) remarks that, at first, “the real negative force of μή was in abeyance,” and , 157) argued that “μή followed by οὐ presents no difficulty grammatically, since οὐ simply negates the verb.” Both of these remarks miss the fact that the typical construction with the subjunctive is a single μή. They also run into the obvious problem of not having enough verbal antecedents to prove the grammatical delineation (e.g., καὶ θρὶξ ἐκ τῆς κεφαλῆς ὑμῶν οὐ μὴ ἀπόληται in Luke 21.18).

50

Moorhouse (1959, 39–40) suggests that in time, “considerations of euphony (avoidance of hiatus) might be sufficient to make one form or the other preferable.” Note also Porter (1992, 281): “The use of the negatives in the NT represents a growing strength of the negative μή, as it encroaches on areas previously reserved for οὐ.” , 108) is here a bit simplistic, but accurate enough for pedagogical purposes: “The words οὐ and μή are synonymous in meaning, but they are typically used in different contexts.”

51

The predominance of this phenomenon being true is referred to as “Blass’s Canon” in Moulton (1906, 170). But note Verbrugge (2013, 586): “I fully recognize there is no hard-and-fast rule about the mood of a verb and use of οὐ versus μή.” Exceptions to the Canon are noted in .

52

On the origin of μή from Sanskrit see , 324).

53

This question may be compared to the question regarding whether the grammar of Greek verbs specifies a particular time referent, or whether “deponent” middle verbs are truly middle in voice. In either case, the question is whether the grammar and syntax specify (or “contain”) a semantic dimension in the communicative act ().

54

Hence the possibility of interchangeability mentioned earlier in , 59).

55

Cf. , 468), though his focus is more on mood than aspect.

56

The temporal emphasis in the current English word “never” seems usually to be in abeyance. Perhaps it was earlier focused on temporal duration but more recently has become synonymous with temporal–neutral absolutes (“no!” “by no means!” “absolutely not!” etc.).

57

Hence the use of οὐκέτι οὐ μὴ in Mark 14.25.

58

Note van der Wouden’s point was pressed in with regard to the semantic significance of the hapax legomenon αὐθεντέω in 1 Tim 2.12. Some have claimed that this term was used for no particular reason (i.e., being synonymous with similar terms like ἐξουσιάζω), but given the rarity of the term and its many potential substitutes, Hübner demonstrates that this conclusion is untenable and that αὐθεντέω has nuance.

59

It is not as if the context is making this determination. Matt 25, for example, takes place with the sudden appearance of the bridegroom and the urgent necessity of lamp oil. Many English translations (therefore?) render οὐ μὴ emphatically (e.g., “No!” in NIV).

60

For a refutation of the idea that οὐ μὴ is nonemphatic in the LXX, see , 95–96).

61

Register is typically determined by occupation and social context, while diatype is associated with social purpose (). Diatype can be broken down into field (subject matter), tenor (the relationship of participants), and mode (form of delivery). Similarly, linguists have defined “register” in many ways besides occupational or specialized language. For example, standard 12620 of the International Organization for Standardization lists the registers as bench-level, dialect, facetious, formal, in-house, ironic, neutral, slang, taboo, technical, and vulgar.

62

Among the 660 occurrences in Rahlfs’s LXX (642 + 18 οὐ γὰρ μὴ constructions), it is absent from Judith, Psalm 151, 3–4 Maccabees, Prayer of Manasseh, Obadiah, and Haggai.

63

asserts that prophetic register—which is the primary soil of οὐ μὴ in both septuagintal and New Testament usage—is also frequently considered formal or “solemn.” Such prophetic contexts can appear in virtually any literary and cultural context.

64

This may explain some commentators’ note of emphasis the rare times when Paul does use it, e.g., Morris (1988, 200); , 389).

65

This article has not addressed idiolect, where one would ask if the use of οὐ μὴ might be attributed to a particular speaker or author’s communicative style (e.g., οὐ μὴ is watered down for some authors—namely those who overuse the expression). Performing a study on the frequency of οὐ μὴ compared and regressed against regular negation (separate occurrences of οὐ and μὴ) of the same author might provide this indication of an author’s habitual, redundant use of οὐ μὴ. However, I performed this brief experiment with multiple regression models of Pauline and Lukan writings, and nothing statistically significant emerged.

66

Special thanks to Laurence Horn (Yale University) and Jeff Miller (Milligan College) for their invaluable and gracious feedback during the development of this article; to Michael Holmes (Bethel University) for pushing me in the direction of linguistics on this topic during breakfast at SBL 2016; and to Wayne Kobes (Dordt University), who taught me undergraduate Greek, where I first began to ask the question for this article.

References

Ballantine

William

. 1897. “Negative Futures in the Greek New Testament.” The American Journal of Philology 18: 453–61.

Barrett

Charles K.

2004. A Critical and Exegetical Commentary on the Acts of the Apostles. Vol. 1. Preliminary Introduction and Commentary on Acts 1–14. International Critical Commentary. New York: T&T Clark.

Bauer

Danker

F. W.

Arndt

W. F.

Gingrich

F. W.

2000. A Greek-English Lexicon of the New Testament and Other Early Christian Literature. 3rd edition. Chicago: University of Chicago Press.

Blass

Friedrich

Debrunner

Albert

. 1961. A Greek Grammar of the New Testament and Other Early Christian Literature. Translated and revised by Funk

Robert W.

Chicago: University of Chicago Press.

Bock

Darrell

. 2007. Acts. Baker Exegetical Commentary. Grand Rapids: Baker Academic.

Bosanquet

Bernard

. 1888. Logic: Or, The Morphology of Knowledge. Vol. 1. Oxford: Clarendon.

Boyer

James

. 1986. “The Classification of Subjunctives: A Statistical Study.” Grace Theological Journal 7: 3–19.

Büchner

Dirk

. 2017. The SBL Commentary on the Septuagint: An Introduction. Atlanta: SBL Press.

Burton

Ernest

. 1900. Syntax of the Moods and Tenses in New Testament Greek. Chicago: University of Chicago Press.

10.

Campbell

Constantine

. 2008. Verbal Aspect and Non-Indicative Verbs. New York: Peter Lang.

11.

Campbell

Constantine

. 2015. Advances in the Study of Greek. Grand Rapids: Zondervan.

12.

Carroll

John

. 2012. Luke: A Commentary. New Testament Library. Louisville: Westminster John Knox.

13.

Dahl

Östen

. 2010. “Typology of Negation.” Pages 9–38 in The Expression of Negation. Edited by Horn

Laurence

. Berlin: Mouton de Gruyter.

14.

Decker

Rodney

. 2015. Reading Koine Greek. Grand Rapids: Baker Academic.

15.

Dhont

Marieke

. 2017. Style and Context of Old Greek Job. Boston: Brill.

16.

Fee

Gordon

. 1987. The First Epistle to the Corinthians. New International Commentary on the New Testament. Grand Rapids: Eerdmans.

17.

France

R. T.

2002. The Gospel of Mark: A Commentary on the Greek Text. New International Greek Testament Commentary. Grand Rapids: Eerdmans.

18.

Giannakidou

Anastasia

. 2012. “The Landscape of Greek Quantifiers.” Pages 285–346 in The Handbook of Quantifiers in Natural Language. Edited by Keenan

E. L.

Paperno

Studies in Linguistics and Philosophy 90. Dordrecht: Springer.

19.

Givón

Talmy

. 2001. Syntax: An Introduction. Vol. 1. Amsterdam: Benjamins.

20.

Goodwin

William

. 1890. “The Origin of the Construction οὐ μὴ.” Harvard Studies in Classical Philology 1: 65–76.

21.

Goodwin

William

. 1900. A Greek Grammar. Boston: Ginn and Co.

22.

Green

Lisa

White-Sustaíta

Jessica

. 2015. “The Development of Variation in Child African American English.” Pages 475–91 in The Oxford Handbook of African American Language. Edited by Lanehart

Sonja

. New York: Oxford University Press.

23.

Gregory

Michael

. 1967. “Aspects of Varieties Differentiation.” Journal of Linguistics 3: 177–97.

24.

Holmes

Janet

Wilson

Nick

. 2017. An Introduction to Sociolinguistics. 5th edition. New York: Routledge.

25.

Horn

Laurence

. 1984. “Toward a New Taxonomy for Pragmatic Inference.” Pages 11–42 in Meaning, Form, and Use in Context: Linguistic Applications. Edited by Schiffrin

Washington: Georgetown University Press.

26.

Horn

Laurence

. 2010. “Multiple Negation in English and Other Languages.” Pages 111–48 in The Expression of Negation. Edited by Horn

Laurence

. Berlin: De Gruyter Mouton.

27.

Horn

Laurence

. 2015. “Lie-toe-tease: Double Negatives and Excluded Middles.” Philosophical Studies 174: 1–25.

28.

Horrocks

Geoffrey

. 2014. “Ouk ’Ismen Oudén: Negative Concord and Negative Polarity in the History of Greek.” Journal of Greek Linguistics 14: 43–83.

29.

House

Mark

Robinson

Maurice

. 2012. Analytical Lexicon of New Testament Greek. Peabody, MA: Hendrickson.

30.

Hübner

Jamin Andreas

. 2015. “Revisiting αὐθεντέω in 1 Timothy 2:12: What Do the Extant Data Really Show?” The Journal for the Study of Paul and His Letters 5: 41–70.

31.

Hübner

Jamin Andreas

. 2018. A Concise Greek Grammar. Rapid City, SD: Hills Publishing Group.

32.

Jackson

Rebecca.

2016. “The Pragmatics of Repetition, Emphasis, and Intensification.” PhD dissertation, Salford University.

33.

Jespersen

Otto

. 1917. Negation in English and Other Languages. Copenhagen: A. F. Høst.

34.

Kiparsky

Paul

Condoravdi

Cleo

. 2006. “Tracking Jespersen’s Cycle.” Pages 172–97 in Proceedings of the Second International Conference of Modern Greek Dialects and Linguistic Theory (Mytilene, Greece, 30 September–3 October 2004). Edited by Janse

Mark

Joseph

Brian D.

Ralli

Angela

. Patras: University of Patras.

35.

Labov

William

. 2001. The Principles of Linguistic Change: Social Factors. New York: Blackwell.

36.

Liddell

Henry George

Scott

Robert

. 1996. A Lexicon: Abridged from Liddell and Scott’s Greek–English Lexicon. 9th edition. Oak Harbor, WA: Logos Research Systems, Inc.

37.

Liddell

Henry George

Scott

Robert

Jones

H. S.

1996. A Greek-English Lexicon. 9th ed. with revised supplement. Oxford: Oxford University Press.

38.

Louw

Johannes

Nida

Eugene

. 1988. Greek–English Lexicon of the New Testament. New York: United Bible Societies.

39.

Lust

Johan

Eynikel

Erik

Hauspie

Katrin

. 2003. A Greek–English Lexicon of the Septuagint. Revised edition. Stuttgart: Deutsche Bibelgesellschaft.

40.

Mathewson

David

Emig

Elodie

. 2016. Intermediate Greek Grammar. Grand Rapids: Baker Academic.

41.

Mitiku

Abera

. 2005. “The Use of οὐ μὴ in the New Testament: Emphatic or Mild Negation?” Faith and Mission 22: 85–100.

42.

Montanari

Franco

. 2015. The Brill Dictionary of Ancient Greek. Edited by Madeleine Goh and Chad Schroeder. Boston: Brill.

43.

Moorhouse

A. C.

1959. Studies in Greek Negatives. Cardiff: University of Wales Press.

44.

Morante

Roser

Sporleder

Caroline

. 2012. “Modality and Negation: An Introduction to the Special Issue.” Computational Linguistics 38: 223–60.

45.

Morris

Leon

. 1988. The Epistle to the Romans. Pillar New Testament Commentary. Grand Rapids: Eerdmans.

46.

Moule

C. F. D.

1959. An Idiom Book of New Testament Greek. Cambridge: Cambridge University Press.

47.

Moulton

James

. 1906. A Grammar of New Testament Greek: Prolegomena. New York: T&T Clark.

48.

Moulton

James

Howard

Wilbert

. 1929. A Grammar of New Testament Greek. Vol. 2. Accidence and Word-Formation. Edinburgh: T&T Clark.

49.

Mounce

Bill

. 2009. The Basics of Biblical Greek. 3rd edition. Grand Rapids: Zondervan.

50.

Muraoka

Takamitsu

. 2009. A Greek–English Lexicon of the Septuagint. Walpole: Peeters.

51.

Porter

Stanley

. 1992. Idioms of the Greek New Testament. London: Sheffield Academic Press.

52.

Porter

Stanley

Land

Christopher

. 2019. “Modeling Negation in Ancient Greek.” Filología Neotestamentaria 32: 3–20.

53.

Robertson

A. T.

1914. A Grammar of the Greek New Testament in Light of Historical Research. New York: Hodder and Stoughton.

54.

Silva

Moisés

. 1994. Biblical Words and Their Meaning: An Introduction to Lexical Semantics. Grand Rapids: Zondervan.

55.

Swanson

James

. 1997. Dictionary of Biblical Languages with Semantic Domains: Greek (New Testament). Oak Harbor: Logos Research Systems, Inc.

56.

Swart

Henriëtte de

. 2010. Expression and Interpretation of Negation: An OT Typology. Studies in Natural Language and Linguistic Theory 77. Dordrecht: Springer.

57.

Swetnam

James

. 1998. An Introduction to the Study of the New Testament Greek. Rome: Pontifical Biblical Institute.

58.

Thompson

Francis

. 1898. A Syntax of Attic Greek. London: Longmans.

59.

Verbrugge

Verlyn

. 2013. “Response to Stanley E. Porter.” The Journal of the Evangelical Theological Society 56: 585–87.

60.

Wallace

Daniel

. 1998. Greek Grammar Beyond the Basics. Grand Rapids: Zondervan.

61.

Willis

David

Lucas

Christopher

Breitbarth

Anne

. 2013. “Comparing Diachronies of Negation.” Pages 1–50 in The History of Negation in the Languages of Europe and the Mediterranean. Edited by Willis

David

Lucas

Christopher

Breitbarth

Anne

. New York: Oxford University Press.

62.

Willmott

. 2013. “Negation in the History of Greek.” Pages 299–340 in The History of Negation in the Languages of Europe and the Mediterranean. Edited by David Willis, Christopher Lucas, and Anne Breitbarth. New York: Oxford University Press.

63.

Wouden

Ton van der

. 1997. Negative Contexts: Collocation, Polarity, and Multiple Negation. Routledge Studies in Germanic Linguistics. London: Routledge.