What Did Pilate State in John 19:22? (repost)

[This is a repost (revised a bit) in recognition of Holy Week. Of necessity, it is a bit technical.]

In reading any common English translation of John 19:22, one finds Pilate saying, “What I have written, I have written.”¹ This is certainly not incorrect, yet I have a feeling some readers may not quite comprehend the significance of this statement, in part, because they are unaware of distinctions in English verb tenses. Some may erroneously think “What I wrote, I wrote” conveys the same meaning. In addition, there is a theologically important connotation in the larger context that readers of the English versions would most likely not perceive.

Below is the Greek of Pilates’s quote, under that is its transliteration (exchanging Greek letters for English equivalents), and below that is a corresponding working English translation:

ὃ γέγραφα γέγραφα
ho gegrapha gegrapha
What I have written I have written

The first word is the relative pronoun who/which/what, appropriately translated “what” in this context. Following that are two verbs, the second an exact duplicate of the first. We’ll reserve the final translation until after our investigation. All Greek finite verbs encode person and number (1^st person singular, I; 2^nd person singular, you; 3^rd person singular, he/she/it; etc.), though not grammatical gender (male, female, or neuter). In the above, each verb is 1^st person singular, and they obviously refer to Pilate. Because finite verbs encode person and number, they may form a subject-verb sentence, depending on context. For instance, Jesus’ final word on the cross is tetelestai (John 19:30), translated “It is finished.”

The verb gegrapha is in the perfect tense-form (of graphō), in the active voice and the indicative mood. The indicative is the most common mood—in Greek and English—expressing facts, false statements assumed to be true, false statements as if true, opinions, or questions. The active voice presents the subject performing the action (Jon ate lunch). Comparatively, the passive voice is used when the subject receives the action (Lunch was eaten by Jon).

The precise meaning indicated by the Greek perfect tense-form is in dispute, though there are a number of theories proposed in recent scholarly literature. Specific discussion of any of these theories is not necessary, however, as some of the older Greek grammars address the issues relative to Pilate’s statement, and we can apply them here. While the Greek perfect tense-form has a wide range of applications, for our purposes it is easiest to conceive it as similar to the English present perfect tense. This is what is reflected in all the English translations of John 19:22 at the above hyperlink (have written).

Some Tense Explanations in English

It may be helpful to briefly explain/review the English present perfect as well as a few other English tenses. If you feel like you are sufficiently proficient in the English, you may skip to the next section.

Verbs in the English present perfect tense express past verbal actions that retain some sort of connection to the present. If I were to state I have written over 100 blog posts, you would rightly infer that all 100+ posts are still available for reading on this blog. They are all in a state of having been written, in a state of ‘written-ness’, and available for viewing.

Alternatively, had I constructed the same sentence but in the English simple past tense—I wrote over 100 blog posts—you might infer that the posts were written at some point in the past yet are no longer available for viewing. More on this below.

The difference between these two English tenses can be found in their respective names. The simple past refers to verbal actions in the past with no further implication of present relevance. The “perfect” in present perfect means “complete”, denoting the past (completed, perfected) portion of the verbal action, while “present” in present perfect indicates the verb’s relevance in the present. The “present” portion of the present perfect tense is formed by using the appropriate auxiliary verb for the present tense, matched by person and number: I have (1st person singular), you have (2^nd person singular), she/he/it has (3^rd person singular). The “perfect” (“complete”, past) portion is formed by using the appropriate past participle of the main verb (I have written, you have written, she/he/it has written).

Though the simple past tense does not imply continuing relevance, this does not necessarily mean there is no connection to the present. The simple past is merely silent regarding current relevance. There may or may not be continuing relevance. Further context may (or may not) illuminate.

Comparatively, a verb in the English present perfect always implies something about the present. Thus, the context will determine which one is more suitable. Let’s make a comparison:

I lost my marbles yesterday.
I have lost my marbles! (and I am frantically trying to find them)

In the first instance, reflecting the simple past tense, this is a simple narrative statement, implying no continuing relevance. The second instance reflects the present perfect tense, of course. If it took ten minutes to find my marbles, then that would have been ten minutes I was in a state of having lost my marbles, frantically trying to find them. Upon finding the marbles I would be in a new, much happier (and more lucid) state of having found my marbles. At that point I could exclaim—again using the present perfect tense—“I have found my marbles!”, illustrating this new state.

I could recount the episode by using the past perfect tense: “I had lost my marbles yesterday”. The past perfect indicates a past action that had subsequent relevance in the past following that action (without commenting on present relevance). After finding my marbles yesterday I could have put the two sentences together, stating: “I had lost my marbles, but after a frantic search I have found them!”. Thus, had lost (past perfect) reflects the past action + its subsequent past relevance, while have found reflects the past action + present relevance.

Context will determine which verb tense is best to use:

Francois made dinner last Thursday. He might even make dinner again next month.
Mom has made dinner. (and dinner is now ready to eat)
Myrna has watched all the Die Hard movies three times, and she plans to watch them all again.
Jacob had washed the towels. He has now placed them into the dryer. (they are currently drying)
Johnny has listened to the train coming into the station every day for the past three years.

While the simple past tense of  Francois’ dinner in this context implies this was a one-time only or rare occurrence, the present perfect tense of Mom’s dinner implies that dinner is now ready to eat (and we better do so before it gets cold!). As for Myrna, the present relevance of the present perfect tense is the fact that this apparent Bruce Willis fan enjoys these movies so much she wishes to watch them all yet again. The past perfect tense had washed represents the necessary prerequisite for Jacob’s currently drying towels, the latter implied by the present perfect has placed and the adverb “now”. In the final example, it is reasonable to infer that Johnny will listen to the trains yet again tomorrow, given the daily recurring (“every day”) has listened, i.e., present perfect tense.

With this brief review of a few English verb tenses completed, we are ready to proceed to the Greek.

Some Tense Explanations in Koine  (New Testament) Greek

The idea of a past action with continuing relevance in the present in the English present perfect tense is the primary thrust of the Greek perfect tense-form. Smyth provides a basic definition: “a completed action the effects of which still continue in the present”.² Dana and Mantey use a broader outline:

The significance of the [Ancient Greek] perfect tense in presenting action as having reached its termination and existing in its finished results lies at the basis of its uses. Emphasis, as indicated by context or the meaning of the verb root, may be on either the completion of the action or on its finished results. This possible difference in emphasis lies at the basis of the variation in the uses of the perfect tense.³

Going back to the aforementioned tetelestai, “it is finished” in John 19:30, this is in the perfect tense-form like the twinned verbs in 19:22. Yet here the verb is in the 3^rd person singular and the passive voice. Note that, with the exception of Young’s Literal Translation, common English versions do not read it has been finished. This is because, though the larger context implies completed actions leading up to the culmination point, translators deem that it is the state of completion that is the emphasis in the immediate context: It is finished.⁴

Similarly, in Jesus’ temptation in the wilderness, He answers the Devil’s commands/temptations with gegraptai, “it is written” (Matthew 4:4,7,10, and parallels)—the “it” referring to Scripture. This verb has the same root (graphō) as our subject verse John 19:22, and it is in the same perfect tense-form. The only difference is that here it is a 3^rd person singular in the passive voice. The common rendering “it is written” reflects the primary focus on its state of having been written, not on the fact that Scripture was written. As A. T. Robertson comments on its usage in these contexts, “It was written . . . and still is on record.”⁵ Just like my example of the blog, both the past action and the present results are encoded, but here it is the latter that is understood as the main point: It is written.

Now going back to English for a few moments, let’s say you have a report that needs to be written by 5 o’clock today. At 4:15 your nosy coworker asks if you will meet the deadline. Given that you typically set out to finish a task at least an hour before the deadline (you do, right?), if you express your answer with the verb finish or write, you could say:

I have written the report. (active voice, present perfect)
I have finished the report. (active voice, present perfect)
The report has been written. (passive voice, present perfect)
The report has been finished. (passive voice, present perfect)
The report is written. (passive voice, ___________?)
The report is finished. (passive voice, ___________?)

Though the last two are unquestioningly passive, there is cause to question whether these convey the same meaning as the middle two sentences. To better illustrate, below is a new sentence, stating it in the English present perfect, first in the active voice, then in the passive, followed by a sentence in the same format as the last two above.

Sally has written the book.
The book has been written by Sally.
The book is written by Sally. (?)

To alleviate any possible confusion, observe that the second sentence contains the same exact verb form (has been written) as the passive example further above regarding the report that was due at work. Note the difference between “has written the book” here and “have written the report” above. The different auxiliary verb (has vs. have) reflects the difference between the 3^rd person (Sally has) and the 1^st person (I have). With this clarified, we’ll resume.

Recall that the English present perfect reflects a completed (perfected) action with relevance in the present. The first two sentences certainly are present perfects. But the third one is an attempt at illustrating the state following the verbal action, in order to focus on the result over against the past action. But is this an accurate way to convey this? Let’s take the same basic sentence just above and put it into the English simple past tense, then the simple present tense, both in answer to the question “Who is the author of this book?”

Sally wrote this book. (active) >> This book was written by Sally. (passive)
Sally writes this book. (active) >> This book is written by Sally. (passive)

Notice how the bolded sentence above is identical to the third one in the previous set. Thus, that sentence, like this one, is in the simple present. Speaking on this specific issue, Smyth writes, “When the [Greek] perfect marks the enduring result rather than the completed act, it may often be translated by the [English] present.”⁶ Did you notice what I did in the previous sentence? I prefaced the Smyth quotation with writes, the English simple present, mirroring the first of the second pair of sentences just above (Sally writes this book.). My objective was to signify the enduring words in Smyth’s grammar book. This is typical English convention—that is, substituting the English simple present for the English present perfect when the verb’s enduring result is the emphasis in a given context, as opposed to the verb’s completed action.

Yet this convention does not work as well in the 1^st and 2^nd person for the English present perfect in the active voice. We can quickly deduce that each sentence on the right below cannot be understood to say the same thing as the one on its left:

I have written the document. >> I write the document. (?)
I think you have written this note! >> I think you write this note! (?)

We are now prepared (finally!) to get back to Pilate’s statement. The first instance of gegrapha is best rendered “I have written”, just like the typical English translation.⁷ This reflects the past action of inscribing JESUS THE NAZARENE, THE KING OF THE JEWS (19:19) as well as the present results reflected in those words as they appear on the sign. We could paraphrase this first part of Pilate’s statement (ho gegrapha):

“What I wrote and currently appears on the sign…”

The enduring relevance of this first verb began at the point at which the inscribing of the inscription had been completed (19:19) and continued until the time Pilate responded to the chief priests (19:21-22). In other words, the duration was relatively short.

But what does the second occurrence of gegrapha mean and how should it be translated? The Jews were unhappy with Pilate’s phrasing of the sign, strongly suggesting he amend it. But Pilate was resolute—he wasn’t going to change it. So, what is the best way to translate this second Greek perfect-tense verb? Strictly speaking, “I have written” is correct, and we may well leave it that way. Yet this second gegrapha focuses on the then-present enduring result of the sign’s ‘written-ness’. Since we cannot use the passive voice (“it is written”, “it stands written”) because this verb is in the active voice, and we have determined that the active voice in the English simple present (“I write”) does not adequately convey the results of a present perfect (“I write”), we may phrase (paraphrase) it something like:

“What I wrote and currently appears on the sign, I shall keep recorded.” (I shall not change)
“What I wrote and currently appears on the sign, I shall retain.” (I shall not change)
“What I wrote and currently appears on the sign, I stand by.” (I shall not change)

The last sentence above may be the best, since it does not use a future auxiliary verb (shall) as do the others. To be sure, the future enduring results are implied, but it’s the present enduring results that are specifically encoded by the Greek perfect tense-form here. Then again, the state of ‘written-ness’ should be understood as remaining unless and until some further action brings about a new state. With none specified in the larger context, it is safe to assume this state will continue on.⁸

Thus, to capture the overall meaning here—though it’s not as pithy as “What I have written, I have written”—I might render John 19:22:

Pilate replied, “What I have written, I stand by.”

Pilate’s inscription JESUS THE NAZARENE, THE KING OF THE JEWS remains written. What does this signify theologically in its context? The next part will elucidate.

______________________________

¹ I am taking the words in the Greek text of John as the words of Pilate, whether or not these reflect his very utterance. The words may well be John’s own rendering in service of a larger theological motif. More on this later.

² Herbert W. Smyth, Greek Grammar, Gordon M. Messing, rev. (Cambridge, MA: Harvard University Press, 1956 [1920]), p 434, § 1945.

³ H. E. Dana & Julius R. Mantey, A Manual Grammar of the Greek New Testament (NY: Macmillan, 1955), p 201 § 184.

⁴ Of course, one must concede that because of the verb root itself—complete, finish—the past necessarily recedes in favor of the state of completion. In other words, it cannot later become ‘unfinished’. Note Marianne Meye Thompson’s objection to this rendering (John: A Commentary, New Testament Library, C. Clifton Black, et al eds. [Louisville, KY: Westminster John Knox, 2015]), “…Jesus’ last words from the cross, tetelestai, ‘It is finished!’ (19:30) surely means ‘it has been accomplished’ or ‘it has been completed’ with reference to completing God’s mission and work” (p 395). I have to agree.

⁵ A. T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research (Nashville, TN: Broadman Press, 1934), p 895.

⁶ Smyth, Grammar, p 434, § 1946.

⁷ Some grammars and commentaries claim this first γέγραφα is “aoristic”, functioning as if an aorist tense, akin to the English simple past wrote. Entailed in this position is that the second γέγραφα is construed as an English present perfect have written, with a focus on the enduring result. Robertson makes a strong case that the perfect never functions ‘aoristically’ in the NT, though it does post-NT era (A Grammar of the GNT, pp 898-902; cf. 895 [β]).

⁸ Yes, of course, the cross was eventually taken down; however, the point here is that the inscription was not altered.

Advertisement

The Holy Spirit as “Restrainer” in 2 Thessalonians 2?

(In this post I had to go into the weeds a bit. My goal is to have trimmed them to a level the reader can see the path I made—to allow the reader to see the forest beyond the weeds. I hope I’ve achieved it.)

Quite a few issues hinder identifying “the restrainer” in 2 Thessalonians 2:6 and 2:7. A fair number of theologians argue for the Holy Spirit in these two verses. And some of these do so based on a questionable pronoun argument. From this shaky foundation the Holy Spirit is imported into the context. This then is used by some as support for the pre-tribulation “Rapture” doctrine.

One shaky foundation upon another.

The Dubious Pronoun Argument

I’ve touched on this dubious pronoun argument elsewhere (Misgendering the Spirit, which has a basis in Another Paraclete?). To briefly identify the issue:

The Greek word for “Spirit” is pneuma, which is neuter in grammatical gender. Because the term is neuter, a literal English translation of the pronouns substituted for pneuma would be “it” or “which” (instead of the personal “whom”). For example, in Acts 5:32, we would translate “we are His witnesses of these words, and so is the Holy Spirit (pneuma), which God gave to those obeying Him [Jesus].” And some deny the personhood of the Holy Spirit simply because pneuma is neuter. But this erroneously imposes English grammar notions upon the Greek.

Some, accepting this error, attempt to counter it by construing (erroneously) a different way to promote the personhood of the Spirit. In John 14:16—16:15 the Holy Spirit is also called another “paraclete” (paraklētos), which is masculine in grammatical gender. In addition, there are a number of masculine pronouns throughout this section. Consequently, some have mistakenly assumed that at least some of these masculine pronouns refer back to pneuma instead of “paraclete”, thus (also erroneously) implying personhood. But it is context that establishes personhood—not pronouns. Moreover, these masculine pronouns all refer back to “paraclete” or to a masculine pronoun substituting for “paraclete”.

An analogy will illustrate the problem with this argument. The term logos, “word”, is a grammatically masculine noun. If we impose English grammar rules upon the Greek here, then every occurrence of logos, would be translated “he” or “him”. This works fine for “the Word” as the pre-incarnate Jesus in John 1:1-14. But it doesn’t work in Mark 4:16, which would be “…whenever they hear the word (ho logos) they immediately receive him (autos) with joy.” This context is about the Gospel message, which should more properly be “it” (not “him”) in English.

The 1977 movie The Car provides a different angle. While the noun “car” in English is “it”—a thing—The Car as portrayed in the movie is a sentient, willful being. So, does The Car have personhood? The Car, which self-drives, certainly has a mind of its own—even mockingly ‘laughing’ via the car horn.¹ Or should we instead say, ‘The Car, who self-drives, certainly has a mind of his own’? ‘A mind of her own’? Do pronouns—whichever we apply—affect the consciousness, the volition of The Car in the movie?

Also consider the many ships named after humans (Queen Elizabeth II; Edmund Fitzgerald). Though we may call them individually “she” or “he” according to name, each is still “it” in English. Unless fictionally personalized as in The Car, none are sentient beings.

Grammar Limits Importing the Holy Spirit into 2 Thessalonians

Those relying on the pronoun argument for 2 Thessalonians 2 do so because there are two different grammatical genders used for “the restrainer”. One is neuter (2:6: to katechon), the other masculine (2:7: ho katechōn). To account for this, Robert L. Thomas argues:

To one familiar with Jesus’ Upper Room Discourse, as Paul undoubtedly was, fluctuation between neuter and masculine recalls how the Holy Spirit is referred to. Either gender is appropriate, depending on whether the speaker (or writer) thinks of natural agreement (masculine because of the Spirit’s personality [Jn 14:26; 15:26; 16:13—14; cf. Robertson, Grammar of the New Testament, 708-9]) or grammatical (neuter because of the noun pneuma…). This identification of the restainer…is most appealing.²

To support his stance, Thomas sources Robertson’s respected grammar. We will critique Robertson’s argument—and by extension Thomas’—with respect to the verses in John after first defining and analyzing “natural [gender] agreement” elsewhere.

“Natural gender” is such that a word’s grammatical gender correlates to the sex/gender inherent in the person the word is referencing. For example, “he” is masculine in grammatical gender, correlating to a male. “Stewardess” implies a woman. Etc.

Below we will challenge the above claim that the Holy Spirit’s “natural” gender is “masculine because of the Spirit’s personality” (personhood).

Defining “Natural [Gender] Agreement”

“Natural [gender] agreement” occurs when a particular word’s gender correlates to the biological sex/gender of the person it directly refers to (“natural gender”) in distinction from the grammatical gender of its antecedent (preceding reference)—the particular word it refers back to. Stated another way, instead of agreeing in grammatical gender with its previous reference, the word agrees with the biological sex/gender of the person (“natural gender”) referred to.

An analogy from English should clarify:

A₁: Though the player grew weary, the athlete was spurred on by the crowd.
B₁: Though the player grew weary, she was spurred on by the crowd.

The antecedent (preceding reference) in each sentence is the player. In A₁ the athlete is unspecified as to gender, just like its antecedent. For the sake of this example, we’ll call both neuter in grammatical gender.  In A₁ there is the usual grammatical gender agreement (the player – the athlete). In B₁ the natural gender of the person (female/feminine, she) to whom the word refers is used instead of one matching the grammatical gender (neuter) of its antecedent. In Greek, this dissimilarity would be a case of applying natural gender agreement in place of grammatical gender agreement—a grammatical mismatch.

This sort of thing is also called “construction according to sense”, or constructio ad sensum.

From Mark’s Gospel, Robertson finds an example of natural gender agreement over the grammatical:

In Mk. 5:41 αὐτῇ [autē̦, her] follows the natural gender of παιδίον [paidion, child] rather than the grammatical.³

The applicable portion of the verse is “Taking the child (paidion, neuter) by the hand, He said to her (autē̦, feminine)…” Alternatively, Mark could have chosen the neuter form of the pronoun (autō̦⁴) for grammatical gender agreement between the two words. By choosing “her” (autē̦), he followed “natural gender” instead. Mark went from the general “child” (paidion, neuter) to the more specific “her” (autē̦, feminine)—a grammatical mismatch.

Let’s look more closely at this. In the larger context, Jesus heals Jairus’ daughter. The feminine thygatēr, “daughter”, is used for her in 5:35. Thus, we already knew the child is female rather than male. But the intervening references, up to and including 5:41 (39, 40, 41), all use the more general neuter paidion, “child” as a synonym. Here is the chain of referents:

thygatēr (5:35) > paidion (5:39) > paidion, paidion (5:40) > paidion (5:41) > autē (5:41)
daughter (5:35) > child (5:39) > child, child (5:40) > child (5:41) > her (5:41)

Thus, while 5:41 clearly is an instance of “natural [gender] agreement” over grammatical agreement, the natural gender is determinable by the larger context by a preceding reference.⁵

An Evil Pneuma Has Masculine “Natural Gender”?

For our purposes here, Mark 9:20 provides a more fitting example. Its context contains “spirit”, pneuma. Robertson opines that this is another example of natural gender usage:

“…surely this is…treating πνεῦμα [pneuma] as masculine (natural gender).”⁶

But what makes a malevolent spirit inherently masculine in “natural gender”? Is a spirit male?⁷ Let’s scrutinize.

In the larger context beginning at Mark 9:17, a particular pneuma possesses a man’s son, preventing him from speaking. In 9:18 a neuter pronoun appropriately correlates to pneuma, its antecedent: “…I asked your disciples to cast it (auto) out…”. Correspondingly, in 9:28 the disciples asked Jesus, “Why weren’t we able to cast it (auto) out?”

Yet 9:20 causes a bit of confusion. It contains four different uses of the masculine ‘personal’ pronoun (auton, “him”). Related to this is a grammatical anomaly: the masculine pronoun encoded in the participle “saw” (he). This will take a bit to sort out:

ēnegkan auton pros auton. kai idōn auton to pneuma euthys synesparaxen auton…
They-brought him to Him. And [when] he-saw Him, the spirit immediately convulsed him…
They-brought the son to Jesus. And when he-saw Jesus, the spirit promptly convulsed the son…

The proper correspondent for the masculine “he-saw” in this sentence would be “the spirit” (to pneuma). But, of course, pneuma is neuter instead of masculine.⁸ The late Rodney Decker explains: “As an adverbial participle idōn [he-saw] should agree [in gender] with the subject of the main verb (pneuma), but this is constructio ad sensum.”⁹ Decker then refers to Ezra Gould, whom I’ll quote in larger context:

[S]ince the action of the verb [convulsed] belongs to the spirit, and is occasioned by the action denoted by the participle [saw], it would be the spirit which is described as having seen Jesus. But [the spirit] does this with the eyes of [the son], and hence the masculine form of the participle.

In all these stories [exorcisms], the man and the evil spirit get mixed up in this way. The outward acts belong to the man, but the informing spirit is sometimes that of the man, and sometimes the evil spirit.¹⁰

In other words, though there is a grammatical gender mismatch, Gould thinks this is due to the mixing of the spirit (neuter) and the boy (masculine). Since the spirit (pneuma, neuter) acts through the son (huios, masculine), a masculine-gendered participle represents the son as the spirit’s ocular vehicle here. The possessing spirit sees through the eyes of the possessed son.

Sometimes the person acts apart from the demon. Sometimes the demon acts through the person. From the evidence thus far, one could consider this a case of “natural gender” agreement (constructio ad sensum) correlated to the son (son/spirit), not the spirit.

Mark 1:23-26 provides a good comparison. The translation below is as ‘literal’ (formally equivalent) as possible. Subscripted brackets identify grammatical number (singular_[sg] or plural_[pl]) and gender (masculine_[ms] or neuter_[nt]) where necessary for analysis:

1:₂₃ Suddenly, in their synagogue was a man with an unclean spirit! He cried-out, ₂₄ saying_{[ms sg]}, “What are we_[pl] to You, Jesus the Nazarene? Have You come to ruin us_[pl]? I-know_[sg] who You are: The Holy One of God.” ₂₅ But Jesus rebuked him/it_{[ms/nt sg]} [autō̦], saying, “You_[sg] be-quiet and you_[sg] come-out of-him [autou]!” ₂₆ And the unclean spirit, [it-]violently-convulsing_{[nt sg]} him and [it-]shouting_{[nt sg]} with a loud voice, came-out of-him [autou].

Like 9:20 above, the possessing pneuma acts through its chosen vessel. This is grammatically shown by Mark’s use of the masculine singular “he”, instead of the neuter in 1:23/24. Note that the spirit self-references with the plural “us” in each of the two questions posed in 1:24. Yet when the unclean spirit reveals that it knows Jesus’ identity, it reverts back to the singular (I-know)!¹¹ In 1:25 Jesus addresses the pneuma either (a) through the man [him, ms], or (b) directly [it, nt], in commanding exorcism. The better of the two would be (b), for it makes the most sense in this context (see next paragraph). In this way, the pronoun refers to the spirit in distinction from its host.¹² And finally, in 1:26 the unclean spirit (nt sg) comes out of the man (autou), freeing him.¹³

Though there are a number of interpretive options, we should reject any notions that the plural indicates more than one spirit possessing the man.¹⁴ Jesus specifically exorcised only one demon. In view of the grammatical evidence of 9:20, at minimum we should harmonize the plural (1:24) with the masculine singular “he”, such that we construe the plural “us” to include the possessed man along with the unclean spirit.¹⁵ Thus, the “I” in 1:24 comes from the spirit, which Jesus rebukes in 1:25. Surely the possessing spirit is concerned Jesus has come to exorcise him it from his its chosen host.¹⁶ And Jesus promptly does.

If we understand the grammatical anomaly of 9:20 through the contextual lens of the grammar in 1:23-26, Gould’s above comments fall right into line.

In the account of “Legion” in Mark 5 we find a similar thing. Once the man with an unclean spirit is introduced (5:2), there seems to be no differentiation between the acts of the man and the acts of the possessing spirit. These verses, again, appear to reflect the spirit’s actions through the man:

5:₆ Seeing_{[ms sg]} Jesus from afar, he-ran and bowed down to-him. ₇ In a loud voice he-was-crying-out, saying_{[ms sg]}, “What am I to You, Jesus, Son of the Most High God? I implore you by God: Don’t torment me!” ₈ For [Jesus]-had-said to-him/it_{[ms/nt sg]}, “You-come-out_[sg] of the man unclean spirit!” ₉ And [Jesus]-had-asked him_{[ms sg]}, “What is your name?” He-replied_[sg] to-Him[Jesus], “Legion is my_[sg] name_[sg], for we-are_[pl] many.” ₁₀ Then he/it-began-begging_[sg] Him further that [Jesus] would not send them_{[nt pl]} out of the region.

₁₁ Nearby, on the hillside, was a large herd of pigs grazing. ₁₂ So they-began-begging_[pl] Him, saying_{[ms pl]}, “Send_[pl] us_[pl] into the pigs so that we-may-enter_[pl] them!” ₁₃ And [Jesus]-permitted them_{[nt pl]}. So, after coming-out_{[nt pl]}, the unclean spirits entered_[pl] into the pigs; and, the herd of pigs—about two-thousand—rushed down the steep bank into the lake. And they were drowned in the lake.

Though the actor in 5:6 is the man, in reality it is the possessing spirit acting through the man, as evidenced by its words in 5:7. In 5:8 where Jesus starts exorcising the spirit through the possessed man, the Gospel writer uses a pronoun (autō̦) that could be either masculine or neuter. We might expect neuter in order to grammatically agree with pneuma, as found at the end of this verse.¹⁷ This would concur with the usage in 1:25-26. And this seems the best understanding here.

In 5:9 Mark records Jesus having asked his (auton, ms) name—apparently to the spirit through the man (or to the man with the spirit answering)—and the unclean spirit replied, “My name is ’Legion’, for we are many.” This plurality of unclean spirits then (5:10) implores Jesus to not send them (auta, nt pl) away.¹⁸ At first blush the masculine singular at the beginning of 5:10 appears to contradict the analysis above; however, the narrator is describing the singular possessed man’s actions spoken by the one spirit representing them all. It is paralleled with 5:9 in that the narrator records a singular spirit (through the man) explaining its collective existence in the man (“Legion”).

Next, in 5:12 Mark records things from the plurality of spirits’ perspective (“They began begging”). Once again the Gospel writer prefaces their statement with a masculine participle (“saying”), but this time he uses the plural to signify the plurality of demons speaking through the man.¹⁹ Evidence to support this position finds itself in the next verse: “Jesus permits them” (nt pl). Further, we can deduce that when the spirit is self-referencing, the neuter should be understood (“me” and “I” in 5:7; “my” and “we” in 5:9; “us” and “we” in 5:12). This also assists in confirming the analysis above that the sequence “us”, “us”, “I” in 1:24 refers to the pneuma/possessed man in the “us” as compared to the pneuma only in the “I”.

When prefacing direct speech—except before words commanding exorcism—Mark uses the masculine to indicate pneuma/possessed person. By contrast, in narration describing the scene the Gospel writer uses, or implies, the neuter (5:10, 12, 13), to distinguish the spirit(s) from the possessed man.

Therefore, all the above best explains the example in 9:20 above. In narration, acts of the pneuma independent of the possessed person are expressed by the neuter. And the neuter is used when the spirit is referenced by another in distinction from the possessed person (9:18, 28). Therefore, it could well be argued that grammatical gender and “natural gender” are the same for pneuma—neuter. Or, perhaps better, that a pneuma does not have “natural gender” at all. To add strength to this position, we’ll look at one final account of demon possession.

The possessed slave-girl in Acts 16:16-18 provides a great comparison and contrast.²⁰ Luke records the girl as speaking what are obviously words coming from the possessing spirit (16:17).²¹ The gender of both the pronoun (hautē, “This one” [slave-girl]) and the participle preceding the speech (legousa, “[she]-was-saying”) is feminine, but the words certainly originate from the pneuma. In other words, the feminine grammatical gender is apparently due to the biological sex/gender of the spirit’s mouthpiece—its verbalizing vehicle.

Thus, harmonizing these accounts with respect to gender—and applying a bit of Occam’s razor—one can make a strong argument that none of these accounts of spirits/demons contain examples of “natural gender” as pertaining to pneuma. That is, pneuma remains neuter. When a masculine or feminine referent is seemingly used for pneuma this merely indicates the biological sex of the possessed person. These examples also cast doubt on any notions of “natural gender” for spirit beings generally. I think we must be very careful not to impose human gender ideas upon spirit beings.

The Holy Spirit Has Masculine “Natural Gender”?

Finally, we shall directly engage Robertson’s argument that the Holy Spirit has implied masculine “natural gender”. He asserts, “Two passages in John call for remark, inasmuch as they bear on the personality of the Holy Spirit.”²² The first part refers to 14:26, the second 16:13. In the following we shall see that English grammar notions inform part of his argument on the Greek, which is then contended to indicate the Spirit’s personhood (“personality”).

This is the first part of the grammarian’s argument (Greek replaced with English transliteration):

In 14:26…ho de paraklētos, to pneuma to hagion, ho pempsei ho patēr en tō̧ onomati mou, ekeinos hymas didaxei, the relative [pronoun] ho follows the grammatical gender of pneuma. Ekeinos (“That One”), however, skips over pneuma and reverts to the gender of paraklētos.²³

Essentially, he is correct, which I will unpack below. First, here’s biblical context for his argument:

14:₂₆ But the paraklētos_[ms]—the Holy Spirit_[nt], Which_[nt] the Father will send in My name—That One_[ms] will teach you all things, and remind you of everything I told you.

Robertson is right: “That One” skips over “the Holy Spirit” to agree in gender with paraklētos (advocate, counselor, helper, comforter). This is precisely why I prefer to separate the middle clause by em dashes (—). The neuter relative pronoun “Which” (hos) correlates to “Spirit” (pneuma), while the masculine demonstrative pronoun (ekeinos) “That One” refers to paraklētos. This is grammatical agreement. There is no reason to press this further. However, Robertson uses this as background for the second part of his argument:

In 16:13 a more striking example occurs, hotan de elthȩ̄ ekeinos, to pneuma tēs alētheias. Here one has to go back six lines to ekeinos again and seven to paraklētos. It is more evident therefore in this passage that John is insisting on the personality [personhood] of the Holy Spirit, when the grammatical gender so easily called for ekeino_[nt] [ED: nt instead of ms].²⁴

With respect to the gargantuan effort evident in Robertson’s mammoth grammar, his conclusion “when the grammatical gender so easily called for ekeino [nt]” is a non sequitur. The chain of referents follows:

paraklētos (16:7) > ekeinos (16:8) > ekeinos (16:13)
paraklētos (16:7) > That One (16:8) > That One (16:13)

After providing a brief synopsis of things to come (16:1-6), Jesus again mentions the paraklētos in this discourse (16:7). “That One” in 16:8 directly refers to paraklētos. The verses following this pronoun (16:9, 10, 11) detail functions of paraklētos; so, the same subject is assumed. 16:12 provides a transition to the next thought. There are neither intervening occurrences of pneuma nor intervening neuter pronouns. Thus, there is no basis to assume a neuter would have been “so easily called for” in place of “That One” in 16:13. In fact, the grammar is against it. Consequently, “That One” in 16:13 directly refers to “That One” in 16:8, which in turn directly refers to paraklētos. Therefore, this is not an implied instance of “natural gender” usage, which would somehow go towards supporting “the personality [personhood] of the Holy Spirit”.

Note that Thomas extends his argument a bit further (“Jn 14:26; 15:26; 16:13—14”). We’ll engage the last reference first, since it is a direct extension of Robertson’s with respect to 16:13. Here’s the text in English:

16:₁₃ But when That One comes—the Spirit of truth—He/It-will-guide you into all truth, for He/It will not speak on His/Its own_[ms/nt], but He/It-will-speak only what He/It–hears. And He/It-will disclose to you things yet to come. ₁₄ That One_[ms] will bring glory to Me_[ms/nt], because He/It–will receive from Me and disclose it to you.

Since Thomas does not develop his argument—in all fairness, his commentary is part of a larger volume, so was necessarily limited—I can only speculate. Thus, I don’t think it very profitable to ‘read his mind’, so to speak. However, to reduce this down to its simplest, there is ambiguity as to whether the verbiage after “the Spirit of truth” pertains to “the Spirit of truth” or, alternatively, to “That One” (paraklētos). If the former is assumed, then the second em dash should be placed at the end of 16:13. But this doesn’t seem to do justice to the context. In any case, with the presence of “That One” (ms) at the beginning of 16:14, it is clear this pronoun is intended to agree with “That One” in 16:13.

But surely if this were integral to Robinson’s argument above, the grammarian would have included it.

Thomas’ final reference is John 15:26. But this is much like 14:26 in which there is a split between paraklētos and “That One”. Thus, Thomas may include it as a parallel reference with Robertson’s 14:26. But I fail to perceive how this helps the cause:

5:₂₆ “When the paraklētos_[ms] comes, Whom_[ms] I will send to you from the Father—the Spirit of truth_[nt], the One Which_[nt] comes forth from the Father—That One_[ms] will testify about Me.”

Daniel B. Wallace, in his grammar, engages this kind of argument.²⁵ After refuting it, he concludes (Greek transliterated):

Although one might argue that the Spirit’s personality [personhood] is in view in these passages, the view must be based on the nature of a paraklētos and the things said about the Comforter, not on any supposed grammatical subtleties. Indeed, it is difficult to find any text in which pneuma is grammatically referred to with the masculine gender.²⁶

In a related footnote, Wallace makes a brief comment about 2 Thessalonians 2:6 and 2:7, noting that “Holy Spirit” is absent from the surrounding context.²⁷ Nevertheless, he does think the Spirit is a possible referent for these two verses, though any argument for this cannot rely on the erroneous claim that masculine is the “natural gender” of pneuma.²⁸

Concluding Thoughts

Pneuma is neuter, whether in reference to the Holy Spirit or an unclean spirit.

Yet, isn’t each individual pneuma in the Mark and Acts accounts above in actual fact a sentient, conscious being? Yes, of course. Though each demon (pneuma) uses its chosen vessel as its actor, each pneuma has its own thoughts and will in distinction from its possessed victim. Each is cast out against its will. And doesn’t such sentience necessarily indicate personhood? And shouldn’t this be somewhat analogous to the Holy Spirit? That is, if an unclean spirit is a person in and of itself, how much more should we consider the Holy Spirit to be a ‘Person’ in and of Itself/Himself, regardless of pronoun usage?

We might say “Sally discerned an evil spirt at that meeting” but we understand that it was the Holy Spirit working through her. The difference, of course, is that a Holy Spirit indwelt Christian is not possessed. While an unclean spirit imposes its own will upon its victim, the Holy Spirit allows the indwelt believer a choice. The believer retains complete control yet can allow the Spirit to work through by submitting to His/Its leading. A Holy Spirit indwelt person can choose either to be led by the flesh or led by the Spirit (Galatians 5:13-26).

Too many times it seems theological motives override grammar and context in a misguided effort to bolster a particular position. Sometimes it is needless, for the doctrine in question is secured by other means elsewhere in Scripture. I’m inclined to believe these are motivated by good intentions. I might call these ‘over-interpretations’ and ‘over-apologetics’.

There are other ways to argue for the Holy Spirit as “the restrainer” in 2 Thessalonians 2. From my own research thus far, however, I don’t find them persuasive. If the Holy Spirit is deemed not to be “the restrainer”, how might that impact the pre-tribulation “Rapture” doctrine? And how might that impact your day to day life?

__________________________

¹ Trivia: the horn blast sequence is the same each time, which is intended to indicate “X” in Morse code: dash-dot-dot-dash. (Technically, the second “dash” is too long—should be the same length as the first—and the space between the first “dash” and the “dots” is a bit too long.)
² Robert L. Thomas, 2 Thessalonians, The Expositor’s Bible Commentary, Vol. 12, Ephesians ~ Philemon; Rev. Ed., D. Garland & T. Longman, Gen. Eds. (Grand Rapids, MI: Zondervan, 2006), pp 470-71; underscore added. (A typographical error in the page number references to Robertson’s Grammar [“208-9”] was corrected.) The bracketed portion was moved from the second set of parentheses to the first, given that it seems to be more accurately apply there (“masculine because of the Spirit’s personality”), which becomes evident below when Robertson is critiqued on this.
³ A. T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research, 4^th ed. (Nashville, TN: Broadman Press, 1934), p 684.
⁴ But, then again, since the neuter is the same as the masculine in the dative (and genitive), this could possibly have caused confusion!
⁵ The original reference to the girl comes in 5:23, in which she is introduced by thygatrion (little daughter)—the diminutive form of thygatēr—and this term is neuter though clearly referencing a little girl by its definition. This illustrates that even when the very definition of a term indicates the sex/gender of the person described by the term this does not necessarily mean it will also correlate to the person’s “natural gender”.
⁶ Robertson, Grammar, p 436.
⁷ Bear in mind that Robertson’s work was written in the early 1900s, well before any modern ideas of gender.
⁸ On the surface, this is a grammatical error. A rough equivalent in English would be, “When the boy saw Jesus, the woman…” in which ‘the woman’ is the same person as ‘the boy’ at the beginning. But I think Mark was quite purposeful here. In light of his previous accounts (1:23-26; 5:2-13)—which we’ll investigate below—the Gospel writer intended to more sharply show the distinction between the possessing spirt and its host.
⁹ Rodney J. Decker, Mark 9—16, Baylor Handbook on the Greek New Testament (Waco, TX: Baylor UP, 2014), p 17 (Greek transliterated). I very much respect the late Decker, recalling his words from the Introduction in the BHGNT here explaining his selection of commentaries sourced for his two-part volume: “My choice to engage these writers beginning with the old ICC volume by Gould (1896)…is an attempt to give writers their due. Sometimes more recent commentators are simply a collection of snippets from older works, with or without credit” (p xxiv).
¹⁰ Ezra P. Gould, The Gospel According to St. Mark, A Critical and Exegetical Commentary; ed. Samuel Rolles Driver, Alfred Plummer, and Charles A. Briggs; Accordance electronic ed. (Edinburgh: T. & T. Clark, 1896), paragraph 5185 (at 9:20). Craig A. Evans, in his WBC (Mark 8:27—16:20, Word Biblical Commentary [Nashville, TN: Thomas Nelson, 2001]), disappointingly, at best merely indirectly addresses the grammatical issue, if at all: “The spirit is said to have seen Jesus, and it is the spirit that is said to have convulsed the boy. The symptoms may have been those usually associated with epilepsy, but the Markan evangelist makes it clear that it was an evil spirit, something distinct from the boy himself, that caused the illness” (p 52, emphasis added). But even this is better than the others consulted which didn’t mention the grammatical issue in any manner.
¹¹ Gender is not specifically expressed here, but neuter is implied from its use in both participles in 1:26.
¹² Interestingly, in my software the NA²⁸ is tagged neuter, while the GNT is tagged masculine. There is no textual variant here.
¹³ Though autou is ms/nt in both 1:25 and 1:26, these obviously refer to the man as distinct from the unclean spirit.
¹⁴ Such a faulty notion can come from the plural “the unclean spirits” (to pneuma to akathartos) in 1:28, but this is unnecessary. An exclamation in the plural for one singular surprising action is not uncommon in English: Upon discovering your seemingly non-mechanically-inclined neighbor had replaced the brake pads on his car you respond, “I didn’t know you worked on cars!”
¹⁵ Contra, e.g., Robert H. Stein (Mark, Baker Exegetical Commentary on the Greek New Testament {BECNT} [Grand Rapids, MI: Baker Academic, 2008]), who specifically asserts, “The ‘us’ refers not to the man and the unclean spirit, but to unclean spirits as a group” (p 87). Couldn’t it be both/and? See next note.
¹⁶ This comports with the local context here. Thus, the verb “destroy” (apollumi)—rendered “ruin” here—in 1:24 is likely intended polysemy. On one level the possessing spirit is concerned Jesus will separate it from its host. This is reflected in the second entry in Frederick W. Danker’s The Concise Greek-English Lexicon of the Greek New Testament (Chicago, IL: Chicago UP, 2009): “’experience disconnection or separation’—a. lose with focus on what one has or possesses” (p 47). Though Danker specifically places Mk 1:24 in his first entry (“’cause severe damage’—a. by making ineffective or incapable of functioning destroy”, p 47), I think both apply. More specifically, the latter (destroy) applies both singly to the possessing spirit here and collectively for all demons in the new eschatological age Jesus inaugurated. See Robert A. Guelich, Mark 1—8:26, Word Biblical Commentary (Dallas, TX: Word Books, 1989), pp 58-60.
¹⁷ Gould, St. Mark, states: “Only the man had been mentioned before, which would lead us to refer this [ambiguous ms/nt pronoun] to him. But the command is evidently addressed to the demon. This confusion is due to the identification of the two” (para 4631 [at 5:8]; emphasis added).
¹⁸ The Textus Receptus has the masculine plural pronoun rather than the neuter. Gould thinks this is original (para 4640 [at 5:10]). In view of overall context (to include 9:20), I’m inclined to disagree with Gould, though the manuscript evidence is far from definitive—from what little I know, admittedly. Yet it’s curious that neither Metzger nor Comfort (New Testament Text and Translation Commentary) make note of this variant, for there seems legitimate cause to question the proper text. The NA²⁸ (Holger Strutwolf, ed., Novum Testamentum Graece, 28th, Accordance electronic ed. [Stuttgart: Deutsche Bibelgesellschaft, 2012], p 119) lists the following in the apparatus:

αυτους αποστειλη D ƒ¹³ 28. 565. 700. 1424. (ƒ¹ 2542) 𝔐 ¦ αποστειλη αυτους A 579. 1241. ℓ 2211 it ¦ αυτον αποστειλη ℵ L (⸉ K W 892) lat sy^p bo ¦ txt B C Δ (⸉ Θ).

It may not matter much in translation, but it might be worth discussion by specialists.
¹⁹ The opinion of Gould, Mark, is nuanced somewhat differently, “Here the subject changes from the man speaking for the demons to the demons speaking through the man” (para 4646).
²⁰ See I. Howard Marshall, Acts, Tyndale New Testament Commentaries (Downers Grove, IL: Intervarsity Press, 2008), pp 285—87.
²¹ F. F. Bruce (The Acts of the Apostles: Greek Text with Introduction and Commentary; 3^rd rev. ed. [Grand Rapids, MI: Eerdmans, 1990]) sources Plutarch’s De defectu oraculorum (9.414E) which “calls such soothsayers…ventriloquists who uttered words not only apparently, but really, beyond their own control” (p 360). Cf. Darrell L. Bock, Acts, Baker Exegetical Commentary on the Greek New Testament [BECNT] (Grand Rapids, MI: Baker Academic, 2007), pp 535 note 2, 535—37.
²² Robertson, Grammar, pp 708-709. Here is the complete quote in its context, with transliterations in parentheses after the Greek: “Two passages in John call for remark, inasmuch as they bear on the personality of the Holy Spirit. In 14:26, ὁ δὲ παράκλητος, τὸ πνεῦμα τὸ ἅγιον ὃ πέμψει ὁ πατὴρ ἐν τῷ ὀνόματί μου, ἐκεῖνος ὑμᾶς διδάξει (ho de paraklētos, to pneuma to hagion, ho pempsei ho patēr en tō̧ onomati mou, ekeinos hymas didaxei), the relative ὅ (ho) follows the grammatical gender of πνεῦμα (pneuma). Ἐκεῖνος (ekeinos), however, skips over πνεῦμα (pneuma) and reverts to the gender of παράκλητος (paraklētos). In 16:13 a more striking example occurs, ὅταν δὲ ἔλθῃἐκεῖνος, τὸ πνεῦμα τῆς ἀληθείας (hotan de elthȩ̄ ekeinos, to pneuma tēs alētheias). Here one has to go back six lines to ἐκεῖνος (ekeinos) again and seven to παράκλητος (paraklētos). It is more evident therefore in this passage that John is insisting on the personality of the Holy Spirit, when the grammatical gender so easily called for ἐκεῖνο (ekeino) [ED: neuter (instead of masculine)].”
²³ Robertson, Grammar, pp 708-709.
²⁴ Robertson, Grammar, p 709.
²⁵ Daniel B. Wallace, Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament (Grand Rapids, MI: Zondervan, 1996), pp 331, 332,
²⁶ Wallace, Grammar, p 332; cf. p 338.
²⁷ Wallace, Grammar, p 332, ftnt 44.
²⁸ Wallace, Grammar, p 332. Wallace does not, however, advance any sort of argument.