Does the Gospel of Mark use Bad Grammar?

by Richard Peachey

Recently I came across Robert H. Stein’s The Synoptic Problem: An Introduction (1987), which explores the similarities and differences among the three “synoptic” gospels, Matthew, Mark, and Luke. Following an opening chapter treating “the literary interdependence of the synoptic gospels,” Chapter 2 of the book argues for the historical priority of Mark’s gospel. Stein’s first argument is titled “Mark’s shortness: the argument from length.” His second argument, which is the one that concerns me here, is labeled “Mark’s poorer writing style: the argument from grammar.”

Stein speaks of “the inferior quality of Markan grammar,” including “matters of vocabulary, style, idiom, and sentence construction.” According to Stein, “Mark possessed lesser literary skills” whereas the gospels by Matthew and Luke “contain a more polished and improved literary form.” Stein goes on to provide six examples of Mark’s alleged inferiority (pp. 52-54), each of which deserves refutation.

Stein’s first example of “inferior quality” involves Mark 10:20, in which the rich young man answers Jesus, in Stein’s rendering, “All these I have observed [ephylazamēn] from my youth.” Stein then comments,

The parallels in Matthew 19:20 and Luke 18:21 change the verb to ephylaza. Although this is not evident in the English translations, Mark has used an incorrect form of the verb. He has used an aorist middle, which Matthew and Luke have changed to the correct aorist active.

Let it be noted first of all that Stein has made an error in transliterating these forms of the verb phylassō (“to guard, keep, observe”) from Greek to English. Both verbs actually contain ‘x‘ where Stein has written ‘z.’

Second, and more importantly, it is not “incorrect” to use the middle voice of this verb in this way — in fact, it is thus used repeatedly in the Septuagint, the pre-Christian Greek translation of the Hebrew Scriptures. As Thayer’s lexicon notes, the middle voice of this verb with the meaning “to keep, observe” is “a usage foreign to Greek writers but very frequent in the Septuagint.”

In Exodus 12:17, in a context of keeping a command of God (just as in Mark 10:20), we read, kai phylaxesthe tēn entolēn tautēn (“and you shall keep this command”). This example contains the future indicative middle of the verb phylassō — a form also found in many other places in the Septuagint (e.g., Exodus 31:13,14; Leviticus 8:35; 18:5,26; 19:3,19,30,37; 20:8; 25:18; 26:2; Deuteronomy 4:6; 5:1,32; 8:1; 11:8,32; 26:16; 29:8). Other instances of this verb in the middle voice are seen in Leviticus 20:22 (aorist imperative middle) and 26:3 (present subjunctive middle).

From all these examples it would seem that either the rich young man patterned his words after the Scriptural (Septuagintal) usage or Mark emulated that usage when translating the young man’s words. Either way, the verb form is established usage rather than “incorrect.”

Stein’s second example of Mark’s “poorer writing style” comes from Mark 2:4, which uses the word krabatton, the  accusative form of Greek krabattos (“pallet”). Stein says, “This term is a slang expression for ‘bed.’” Then he claims, “Matthew and Luke change this term to the more-acceptable ‘bed’ (klinēs) and ‘bed’ (klinidiō).

The problem with this argument is that the term krabattos is used not only by Mark (2:4,9,11,12; 6:55), but also by Luke (in Acts 5:159:33) and John (5:8,9,10,11,12). If Luke and John also used this word, why should Mark alone be criticized for using less-acceptable “slang“?

Furthermore, Mark is well aware of the term klinē — he uses it three times, exactly the same number of times as Luke — see Mark 4:21; 7:4,30; Luke 5:188:1617:34. (Matthew uses the word just twice, in 9:2,6.)

As for the term klinidion, this word is used only by Luke (5:19,24) and not by any other New Testament writer — so it’s unsurprising that Mark did not use it.

Mark’s gospel is thought to have been composed in Rome, likely for a Gentile Roman audience. In that case, it would make good sense for him to use the word krabattos, which is related to the Latin term grabātus, meaning “a camp bed, cot, or pallet.”

To call a word “slang,” with the implication that the word is to be viewed as socially less-acceptable, is obviously something of a subjective judgment. The Greek of the New Testament is koiné, the language of ordinary people, rather than some élitist form of classical Greek. So the use of words considered “slang” by a classically trained scholar — or, according to Liddell & Scott’s lexicon, by one particular second century A.D. writer of Attic (Athenian) Greek — is no firm basis for condemning a New Testament writer.

Stein’s third example of Mark’s “lesser literary skills” is found in Mark 1:12. Stein observes:

“. . . we read that after Jesus’ baptism the Spirit ‘drove” (ekballei) him into the wilderness to be tempted. Matthew 4:1 has ‘Jesus was led up [anēchthē] by the Spirit,’ and Luke 4:1 states that Jesus ‘was led [ēgeto] by the Spirit.’” [Note: anēchthē is an aorist indicative passive form of the Greek verb anagō, and ēgeto is an imperfect indicative passive form of agō.]

Stein then proceeds to contrast Matthew and Luke’s “good Greek” with Mark’s “cruder and more confusing term.”

It’s true that the Markan verb ekballō, like its root ballō, has a wide semantic range. The verb can be used for forceful actions such as “drive out” or “throw out”, as well as for calmer acts such as “send out,” “put out,” “take out,” or “bring out.” The context often helps the translator determine what sort of action is intended, though sometimes a bit of ambiguity may remain.

But it needs to be noted that ekballō is used 28 times in Matthew, as well as 20 times in Luke and 5 times in Acts (also written by Luke), in addition to the verb’s 18 appearances in Mark. Indeed, ekballō is used in quite positive settings in Matthew 9:38; Luke 10:2; John 10:4 — with contexts somewhat reminiscent of Mark 1:12. In light of this, why should Mark alone be charged with using a “cruder and more confusing term“?

As well, it must be recognized that the Greek verb agō, used by Luke (and serving as the root of anagō, used by Matthew), also has a wide semantic range. In some contexts it too can signify forceful actions, and may be translated using such words as “impel,” “drive off,” “conduct with force.” Balz & Schneider’s Exegetical Dictionary of the New Testament states that agō “is frequently used of forced leading” and that there are “24 occurrences” of such usage.

The following underlined verbs are all translations of agō in the English Standard Version:

• “you will be dragged before governors” (Matthew 10:18)

• “they rose up and drove him out of the town and brought him to the brow of the hill . . . so that they could throw him down the cliff” (Luke 4:29)

• “But as for these enemies of mine, who did not want me to reign over them, bring them here and slaughter them before me” (Luke 19:27)

• “Then they seized him and led him away” (Luke 22:54)

Other occurrences of agō in forceful contexts include Mark 13:11; Luke 23:1,32; John 7:4518:13,28; Acts 6:128:32; 9:2,21; 18:12. Matthew’s word, anagō, is also found in at least one forceful context (Acts 12:4). So it seems that Stein’s claim that only anēchthē (anagō) and ēgeto (agō) are “good Greek” while ekballei (ekballō) is the “cruder and more confusing term” is quite disputable.

Stein’s fourth example of Markan inferiority is this:

In Mark 4:41 we find still another example of incorrect grammar in Mark’s use of the singular verb ‘hears’ with the preceding compound subject—’wind and sea.’ Both Matthew and Luke correctly use a plural verb.

Stein is correct that Mark’s verb hypakouei is singular in form, though it should be noted that the word is generally translated “obeys” rather than “hears.”

But there is no need to charge Mark himself with bad grammar: he is reporting what was said by some very frightened disciples of Jesus: “Who then is this that both the wind and the sea obeys (hypakouei) him?” The disciples’ words are prefaced with the narrative statement, “And they feared with great fear, and they were saying one to another. . . .” So could it not be that their speech became muddled due to their state of mind? And that Mark has accurately reported their fearful, ungrammatical words?

Alternatively, perhaps several disciples were speaking at once (“saying one to another,” as Mark reports) in a sequence something like this: “Who then is this that. . .?”—”Even the wind!?”—”Even the sea obeys him!” (Note that Mark’s gospel in the original Greek included no punctuation.)

The Expositor’s Greek Testament (Nicoll) offers this suggestion: “. . . the wind and the sea [are] thought of separately, each a wild lawless element, not given to obeying: even the wind, even the sea, obeys Him!”

Another possibility might be that “the wind and the sea” are treated as a unity in the disciples’ expression — something like “All of nature around us (both the wind and the sea) obeys him.” Rodney J. Decker (A Handbook on the Greek Text: Mark 1-8) favours an explanation of this sort. In a footnote, Stein himself allows for such a possibility in the case of Matthew 6:19 where a compound subject “moth and rust” is followed by a singular verb; he also mentions 1 Corinthians 15:50, where the compound subject “flesh and blood” is followed by a singular verb. In these two cases Stein thinks the linked subjects are “no doubt considered as a unity.” (But he views these two examples as different from Mark 4:41 because the nouns are anarthrous, i.e., without the definite article.)

It’s remarkable that Stein takes no notice of James 5:3, which is almost an exact parallel of Mark 4:41— a double subject, each item with the definite article, followed by a singular verb. (A. T. Robertson, in his magisterial Grammar of the Greek New Testament in the Light of Historical Research, p. 405, calls this “a striking example.”) James 5:3 has ho chrysos hymōn kai ho argyros katiōtai (“the gold of you and the silver has been corroded“), just as Mark 4:41 has ho anemos kai hē thalassa hypakouei (“the wind and the sea obeys“).

Also worth noting is the fact that many other NT texts join a singular verb with a compound subject — for example, Matthew 5:18; 17:3; 27:61; Mark 3:318:27; 13:3; Luke 2:338:19; John 2:2; 3:224:36,53; Acts 5:2916:31; 25:23. It’s true that (as Stein points out) in all those cases the verb precedes the subjects, which is not the case in Mark 4:41. Nonetheless this list of examples provides broader evidence that koiné Greek writers were not particularly outraged by the assigning of a singular verb to a compound subject.

In any case, there exists a variety of reasonable suggestions other than that Mark must be charged with “incorrect grammar.”

Stein’s fifth example of “the inferior quality of Markan grammar” is from Mark 16:6. Stein writes,

Here the angel tells the women, ‘”See the place where they laid him.”‘ In English the grammatical errors by Mark are not evident, but in the Greek text we find two of them. The first involves the verb ‘see.’ This is singular in Greek even though the angel is addressing the women (plural). It should be idete as we find it in Matthew [28:6], and not ide as it is in Mark. Also in Mark the term ‘the place’ is in the nominative instead of the accusative case, i.e., it has a case ending that designates it as the subject of the sentence rather than as the object! Matthew correctly converts this to the accusative and has ton topon instead of Mark’s incorrect ho topos.

The answer to this double-barreled accusation is that Stein appears to have misunderstood the nature of the Greek word ide. Although ide is formally a second person singular aorist imperative form of the verb “to see,” it is used in that way only a few times in the New Testament (see John 1:46; 7:52; 11:34; 20:27; Romans 11:22). Most of its NT occurrences represent ide the interjection/particle rather than ide the imperative verb.

According to Abbott-Smith’s Greek lexicon, p. 212, ide is found in the NT as an “interjection, apart from the construction of the sentence, used where one or many are addressed.” Thayer’s lexicon, p. 296, likewise, states that “in most places in the N.T. it stands out of construction like an interjection, even when many are addressed.” Balz & Schneider’s Exegetical Dictionary of the New Testament states that the original imperative verb “became established as a particle,” and that it “is used when more than one person is addressed,” and that what follows “is in the nominative instead of the accusative.” Rather than an imperative verb, ide is usually a stand-alone exclamation: “See!” “Behold!”

Contrary to Stein’s apparent expectation, there are few New Testament texts where ide is clearly followed by a substantive (noun or adjective) in the accusative with which ide treated as a verb must be construed (such would include John 7:52; 20:27; Romans 11:22, listed above). But ide is clearly followed by a nominative not only in Mark 16:6, but also in Mark 3:3411:21; 13:1,21; John 1:29,36,47; 3:265:1412:1918:2119:14,26,27; and Galatians 5:2.

(In several texts — Matthew 25:25; 26:65; Mark 2:2413:21; John 7:2611:3616:29; 19:4 — ide is not followed by a substantive at all. This reinforces the reality that the word is being used as an interjection rather than as a transitive imperative verb which would call for a direct object to be stated. In Mark 15:35 the noun following ide is indeed in the accusative, but that word is, contextually, clearly the direct object of the transitive verb after it, and definitely not of ide. In Matthew 25:20,22; Mark 15:4; John 11:3 ide is followed by an accusative which is arguably governed by the transitive verb after it; at best these few texts, and these alone, could be claimed to be ambiguous cases.)

As noted previously, Abbott-Smith’s lexicon states that ide may be “used where one or many are addressed.” (The others cited spoke similarly.) That is, the expression is not limited to use with a singular addressee/hearer as seemingly expected by Stein. This is a further indication that ide is being used as an interjection rather than as an imperative verb (in addition to the fact already considered, that ide is often followed by a substantive in the nominative case). In each of the following texts the context makes it quite clear that more than one hearer is being addressed by the person using the word ide: Matthew 26:65; Mark 3:3413:2115:35; John 1:3612:19; 19:4,14; Galatians 5:2.

Galatians 5:2 is an especially important text in this discussion. The verse begins, Ide egō Paulos legō hymin (“See! I Paul am saying to you [plural]”). This statement was written by an eminent trained scholar who could not credibly be charged with being “inferior” in his knowledge of Greek grammar. Yet Paul unambiguously uses ide with the nominative: both egō (“I”) and Paulos (“Paul”) are found in the nominative case. As well, Paul is using ide while speaking to multiple addressees — hymin (“to you [plural]”). Thus Galatians 5:2 provides a powerful and thorough refutation of Stein’s charge that Mark has made “grammatical errors . . . two of them.”

Very interestingly, Stein’s 2008 commentary on Mark’s gospel (Baker Exegetical Commentary on the New Testament: Mark), p. 729, renders the key part of Mark 16:6 like this: “See, the place where they laid him.” The now-inserted comma seems to imply that Stein eventually came to understand ide as an interjection rather than an imperative. He does not, however, make any comment on ide at this point in his commentary, nor is the term listed in his index of Greek words!

Stein’s sixth and last example of how Mark was “not a Greek literary giant” is found in Mark 5:9-10, compared to Luke 8:30-31. Here Stein is concerned with Mark’s “grammatical fluctuations” as the pronouns change from singular to plural, then back to singular, then again to plural, during Jesus’ questioning of the man with the “legion” of demons. Luke, in the two verses cited, makes only one move from singular to plural.

But Stein is being too selective with his evidence! Only three verses earlier, in Luke 8:27, there is a mention of “demons” (plural), following which 8:29 speaks of “the unclean spirit” and “it” and “the demon” (singular). Then in 8:30 Jesus questions “him” (= the man? the demon?) about his name, resulting in the response “Legion” along with Luke’s explanation “for many demons had entered him.” Thereafter the plural is used consistently: “they,” “them,” “the demons.” So as it turns out, Luke also includes “grammatical fluctuations,” moving from plural to singular to plural, with some possible ambiguity about whether the man or the demon(s) is/are being questioned by Jesus, and who is doing the answering.

With those facts in mind, Mark’s presentation is not all that different from Luke’s. He does not give an early reference to the plurality of the demons (as Luke does in 8:27), but then he alternates between singular and plural during Jesus’ questioning more than Luke does. Both of these gospels alternate back and forth, and both seem to leave some ambiguity regarding whom Jesus is addressing and who is doing the answering.

Matthew’s presentation of the incident (8:28-32), which Stein does not discuss, mentions two demon-possessed men rather than one (as in Mark and Luke) and describes demons only in the plural, thus avoiding “grammatical fluctuations” altogether. Does this make Matthew more of a “Greek literary giant” than Luke? Such a conclusion would be upheld by very few, I suspect.

The upshot is this: Matthew, Mark, and Luke present their material in different ways, and each writer makes selections regarding what to include in his presentation. Their choices involve what to focus on, and what level of detail to provide to the reader.

Likely there was one demon-possessed man who was more prominent in the episode than the other man: Matthew elects to mention both, whereas Mark and Luke choose to narrow their report to only the more prominent one. When the legion of demons is being highlighted, the writer uses plural verbs and pronouns, but when either the man or the demonic leader/spokesman is in view (especially in Mark, less so in Luke), singular verbs and pronouns are used. But these differences, while interesting and of some importance, do not involve genuine “grammatical” issues or conflicts.

(Note: in Stein’s 2008 commentary on Mark’s gospel, p. 255, he acknowledges: “When the man is in the forefront, we have the singular. . . . When the demons are in the forefront, we have the plural. . . .” This would seem to be an adequate explanation of Mark’s “fluctuations,” eliminating any need for charges of grammatical impropriety.)

In conclusion, I am aware that the majority of New Testament scholars accept the historical priority of Mark’s gospel. In favour of that position there are undoubtedly some good arguments.

But my contention here is that Mark’s alleged “inferior . . . grammar” and “poorer writing style,” as maintained by Robert H. Stein, does not constitute a good argument for Markan priority.