Now it was Pilate’s custom at the feast to release to the people a prisoner of their choosing.
Berean Standard Bible · CC0
Greek — tap a word ↓
de Kata heortēn apelyen autois hena desmion hon parētounto
The Fallible · Synthetic · Study Bible
The Crowd Chooses Barabbas
Mark 15:6–11 — The Crowd Chooses Barabbas. Each verse below carries the full apparatus: the Berean Standard Bible, the vocalized original (tap any word), and a parsed breakdown of every term transcribed from the interlinear. Synthesized commentary, canonical threads, and the reading of Christ gather at the end, over the whole unit.
Now it was Pilate’s custom at the feast to release to the people a prisoner of their choosing.
Berean Standard Bible · CC0
Greek — tap a word ↓
de Kata heortēn apelyen autois hena desmion hon parētounto
And a man named Barabbas was imprisoned with the rebels who had committed murder during the insurrection.
Berean Standard Bible · CC0
Greek — tap a word ↓
de ho legomenos Barabbas ēn dedemenos meta tōn stasiastōn hoitines pepoiēkeisan phonon en tē stasei
So the crowd went up and began asking Pilate to keep his custom.
Berean Standard Bible · CC0
Greek — tap a word ↓
kai ho ochlos anabas ērxato aiteisthai kathōs epoiei autois
“Do you want me to release to you the King of the Jews?” Pilate asked.
Berean Standard Bible · CC0
Greek — tap a word ↓
Thelete apolysō hymin ton Basilea tōn Ioudaiōn de Ho Pilatos apekrithē autois legōn
For he knew it was out of envy that the chief priests had handed Jesus over.
Berean Standard Bible · CC0
Greek — tap a word ↓
gar eginōsken dia phthonon hoti hoi archiereis paradedōkeisan auton
But the chief priests stirred up the crowd to have him release Barabbas to them instead.
Berean Standard Bible · CC0
Greek — tap a word ↓
de Hoi archiereis aneseisan ton ochlon hina apolysē ton Barabban autois mallon
The verse-by-verse work is done. What follows gathers the whole unit. All three layers below are machine-generated (⚙). Weigh them; they have no authority.
AI synthesis — woven from the public-domain voices above and the original text; generated and fallible.
⚙ Mark builds the amnesty out of grammar. There is no Greek noun for "custom" anywhere in this unit; there are only two imperfects, ἀπέλυεν (he was releasing, v. 6) and ἐποίει (he was doing, v. 8). Ellicott states it exactly: both verbs stand in the tense which implies custom, and he corrects the older English of verse 8 to "as he ever used to do unto them." Where the practice came from is genuinely unknown. Cambridge says so plainly, that the origin of the custom is unknown and that it may have been of Jewish origin continued by the Roman governors from motives of policy, and Cambridge also denies, from the absence of the article in κατὰ ἑορτὴν, that it was confined to Passover. Benson takes the opposite line, tying the release to the feast of deliverance from Egypt and calling it therefore agreeable to the nature of the feast, citing Whitby. Gill records the versional split: the Syriac, Arabic, Persic and Ethiopic read "at every feast," though he holds it was only at Passover. The brief does not adjudicate, and neither will we. What the Greek does settle is that Rome had a habit of letting the crowd name one man, and that the crowd had a habit of asking, παρῃτοῦντο, imperfect middle: they kept begging a prisoner off, for themselves.
⚙ Verse 6 puts one prisoner in the balance; verse 7 fills the other pan. δεδεμένος is a perfect participle, a man bound and still bound, and Ellicott notes that Mark alone preserves the detail that his fellow insurgents were jailed with him. Cambridge places him historically as the leader of one of the numerous insurrections against Roman power in which blood had been shed and apparently some Roman soldiers had been killed; Gill says he had been at the head of a seditious mob. On the name itself the sources agree and stop: Strong's G912 glosses Βαραββᾶς as "son of Abba," Pulpit says it is better written Bar-Abbas and means "son of father," and Cambridge gives both derivations along with the striking manuscript variant at Matthew 27:16 in which the man's full name is Jesus Bar-abbas, a reading Cambridge says is supported by the Armenian and Syriac and cited by Origen. None of the commentators in this brief makes a doctrine of the name. We note only what the page itself does: Mark names Barabbas four times in six verses and never once writes the name Jesus, who is only "him" and "the King of the Jews."
⚙ Pilate's question in verse 9 is grammatically hopeful. Ellicott finds in the Greek form, as in the parallel at John 18:39, a half hope of an affirmative answer, and Cambridge says the people's own request concurred with Pilate's wishes so that he resolved deliberately to give the populace their choice. Pulpit reconstructs the calculation: knowing the priests acted from envy, he reasonably supposed the people, who had experienced so many kindnesses, would want him to live. Benson is the dissenting voice and the harshest: if Pilate made this proposal to preserve the life of Jesus, it is hard to say which he discovers most, his want of justice, or of courage, or of common sense. Verse 10 then supplies the motive as a stated cause, διὰ φθόνον, διά plus accusative. Cambridge specifies its object, envy of the influence He had gained and the favour He had won throughout the land; Pulpit calls envy the low passion that influenced the chief priests; Gill says it was not from any principle of equity and justice, nor from any regard to Caesar. That the prosecution's motive is recorded by the governor who then convicts anyway is the hinge of Mark's account.
⚙ The unit ends on its rarest word. Cambridge observes that ἀνέσεισαν occurs only here and at Luke 23:5, and lays out the force: to shake to and fro, to brandish, to make threatening gestures, to instigate. Pulpit compares Matthew's milder ἔπεισαν, they persuaded, and says Mark's word implies a rousing of their bad passions, agitating them to a blind zeal for his crucifixion. The syntax marks the same reversal: in verse 8 ὁ ὄχλος is nominative and acts, in verse 11 τὸν ὄχλον is accusative and is acted upon. Benson says the priests moved the people to desire the release of Barabbas, though he was a scandalous and outrageous criminal; Gill, borrowing Peter's language from Acts 3:14, says they chose rather to have a murderer granted unto them than the holy and just one. And the Greek sentence closes on μᾶλλον, rather: one word, holding two men.
⚙ Read under Sola Scriptura and offered to be tested: this passage is not primarily about Pilate's weakness or the crowd's fickleness, though both are on the page. It is about a substitution performed in public, with paperwork. Roman law had a slot for exactly one release, and Mark's four uses of ἀπολύω track that slot until a name drops into it. The man who fills it is a convicted insurrectionist and, by Peter's word in Acts 3:14, a murderer; the man left standing is the one whose innocence the presiding magistrate has already perceived, ἐγίνωσκεν, imperfect, while the hearing ran. Scripture elsewhere calls that arrangement an abomination: he who justifies the wicked and condemns the righteous, both alike are detestable to the LORD (Proverbs 17:15). The passage therefore stages the exact injustice God's own law forbids, and Scripture's own answer is that God does at the cross what no judge may do, and does it righteously, because the righteous One consents and bears the guilt himself (Isaiah 53:6). The commentators in this brief do not draw that line; they stop at envy, cowardice and mob psychology, and they are right about all three. Our fallible addition is only this: the transaction Mark records is not merely a miscarriage of justice that God later redeemed, it is the shape of redemption itself, visible in a Roman courtyard before a single nail is driven. Test it against the text.
Reader's note, not Scripture, and offered to be tested: one release was available, and the guilty man walked out under a name that means son of the father.
AI-generated connections. Each carries a verification badge with a recorded basis; contested links are flagged.
The verb Mark chooses in 15:11, ἀνέσεισαν (anaseiō, G383), is almost unique in the New Testament. Cambridge states it outright: the word occurs only here and in the parallel, Luke 23:5. In that parallel the chief priests are the accusers, charging that Jesus stirs up the people. In Mark the same rare verb is predicated of the priests themselves, and Pulpit notes that Mark's word is stronger than Matthew's ἔπεισαν, they persuaded. The Verifier returns Luke 23:5 as the highest-scoring candidate for this unit on exactly this lexeme.
Luke 23:5
basis: Verifier: shared Greek lexeme G383 ἀνασείω, present in only 2 verses in the whole New Testament, both within the passion narrative. A genuinely rare verbal link between two Greek texts, not a quotation of one by the other; the tier records the lexical identity, and G756 ἄρχομαι (in 84 vv) in the same candidate is common and carries no weight.
Every evangelist records the same transaction with the same two words. The Verifier scores Matthew 27:16-17, 27:20-21, 27:26, Luke 23:18 and John 18:40 against this unit on Βαραββᾶς (G912) together with ἀπολύω (G630), the release verb Mark uses four times. Cambridge notes at Matthew 27:16 the manuscript reading "Jesus Bar-abbas," supported by the Armenian and Syriac and cited by Origen, so the choice may have been put to the crowd between two men bearing the same first name. Ellicott adds that Mark alone preserves the detail of the fellow insurgents in the cell.
Matthew 27:16 · Matthew 27:20 · Luke 23:18 · John 18:40
basis: Verifier: shared Greek lexeme G912 Βαραββᾶς (in 10 vv, rare and proper) across Matthew 27:16/27:20, Luke 23:18 and John 18:40, with G630 ἀπολύω (in 61 vv) and G1198 δέσμιος (in 16 vv) supporting. Gospel parallels, not quotation of one another; all four are Greek New Testament texts, so a shared-lexeme basis is legitimate.
Mark writes φόνον (murder, G5408) in 15:7 and φθόνον (envy, G5355) in 15:10. The two nouns differ by one letter, each occurs in only nine verses, and Paul sets them side by side in a single vice list at Romans 1:29. The Verifier returns Romans 1:29 on precisely this pair, and Matthew 27:18 on φθόνος together with παραδίδωμι. Mark's narrative makes Paul's list concrete: the man guilty of φόνος is released, and the men guilty of φθόνος walk free of any charge at all.
Romans 1:29 · Matthew 27:18
basis: Verifier: shared Greek lexemes G5408 φόνος (in 9 vv) and G5355 φθόνος (in 9 vv) with Romans 1:29; G5355 φθόνος plus G3860 παραδίδωμι (in 117 vv) with Matthew 27:18. Both lexemes are rare, but Romans 1:29 neither quotes nor alludes to Mark; the shared vocabulary supports a thematic link only, so the tier is deliberately downgraded from verbal.
⚙ On the one day of the Hebrew year when guilt was publicly transacted, two goats were set before the sanctuary and lots were cast: one was killed, and one was presented alive and let go, carrying the iniquities away (Leviticus 16:7-10, 16:20-22). Mark's scene has the same architecture: two men, one release available, and the sins of the people determining which is which. The inversion is exact and total. In Leviticus the goat that goes free carries the guilt away; in Mark the man who goes free is the guilty one, and the guilt is carried by the man who stays. None of the commentators gathered in this brief draws this comparison. It is offered as a figural reading, not as an exegetical claim about Mark's intent.
Leviticus 16:7-10 · Leviticus 16:20-22
basis: Cross-Testament, Greek to Hebrew, so no shared Strong's lexeme is possible and no verbal tier is available; the Verifier returned no candidate here. The basis is structural correspondence only, a pair of subjects with one release between them on a festival day of atonement. Novel with respect to this brief: Ellicott, Benson, Henry, Barnes, JFB, Poole, Gill, Geneva, Cambridge and Pulpit are all silent on it at these verses.
⚙ The law of Israel names this exact courtroom outcome and calls it detestable: acquitting the guilty and condemning the innocent (Proverbs 17:15), and woe to those who acquit the guilty for a bribe and deny justice to the innocent (Isaiah 5:23). Mark supplies every element of the condemned formula. Pilate has perceived the prosecution's motive, ἐγίνωσκεν, imperfect and ongoing, and Gill records that it was not from any principle of equity and justice. What Pilate's court performs is therefore not an error but the specific act the Scriptures forbid.
Proverbs 17:15 · Isaiah 5:23
basis: Cross-Testament, Greek to Hebrew, so shared Strong's numbers cannot be computed and the tier can never be verbal. The basis is the legal formula itself, acquitting the guilty and condemning the righteous, which Mark's narrative supplies in full. Widely-held as a moral reading; no commentator in this brief cites these verses here.
Gill, commenting on 15:7, points forward to Acts 3:14, where Peter tells the same city that they denied the Holy and Righteous One and asked for a murderer to be granted to them. The verb behind Peter's "asked" belongs to the same family Mark uses of the crowd in 15:8, αἰτεῖσθαι (G154), and "murderer" is cognate with Mark's φόνον (G5408). Gill's phrase at 15:11, "the holy and just one," is lifted straight from that sermon, so his exposition of Mark is already Acts-inflected, and he says so.
Acts 3:14
basis: The Verifier did not return Acts 3:14 among the computed candidates for this unit, so no machine-verified shared-lexeme basis is recorded here. The link rests on Gill's own explicit citation of Acts 3:14 at Mark 15:7, plus cognate vocabulary (αἰτέω / φόνος family). Tiered structural rather than verbal because the shared-lexeme basis is uncomputed and the cognates are not identical lexemes.
Luke's account of the same hours uses the same two rare nouns as Mark 15:7, στάσις (G4714) and φόνος (G5408), in Luke 23:19 and again in the closing summary of Luke 23:25, where Pilate releases the man imprisoned for insurrection and murder and delivers Jesus to their will. Benson notes that insurrection was a crime the Roman governors, and Pilate in particular, were more especially concerned and careful to punish. The Verifier scores both Lukan verses on these lexemes.
Luke 23:19 · Luke 23:25
basis: Verifier: shared Greek lexemes G4714 στάσις (in 9 vv) and G5408 φόνος (in 9 vv) with Luke 23:19, and the same pair plus G630 ἀπολύω (in 61 vv) with Luke 23:25. Both lexemes are rare, but these are Synoptic parallels describing one event rather than one text quoting another, so the link is recorded as thematic.
AI-generated reading; weigh it against the text.
⚙ The mechanism of the atonement is visible here before any theology is spoken over it. One release was available; a convicted murderer received it; the man the magistrate had already perceived to be prosecuted out of envy did not. Isaiah had written that the LORD has laid on Him the iniquity of us all, and that He was pierced for our transgressions (Isaiah 53:5-6); Paul states the exchange as doctrine, that God made Him who knew no sin to be sin for us (2 Corinthians 5:21). Matthew Henry, in the brief, reads the whole block in this key when he calls believers to remember the bonds of the Lord Jesus, as bound with him who was bound for us. The substitutionary reading of the passion is the common inheritance of the church; what this unit adds is that the substitution was executed as a public legal transaction, with a Roman procurator's signature on it.
Isaiah 53:5-6 · 2 Corinthians 5:21 · 1 Peter 3:18
⚙ The sourced facts are narrow and firm: Strong's glosses Βαραββᾶς as "son of Abba"; Pulpit says the name means "son of father"; Cambridge gives both derivations and records the manuscript reading at Matthew 27:16 in which the man is called Jesus Bar-abbas, cited by Origen and supported by the Armenian and Syriac. If that reading stood in the crowd's hearing, Pilate offered a choice between Jesus son of the father and Jesus called Christ. The Son who prayed Abba, Father in Gethsemane (Mark 14:36) was kept, and the man whose by-name said son of the father was let go in his place. We mark this honestly: not one commentator in this brief presses the name this way, and the Jesus Bar-abbas reading is a textual variant, not the received text of Mark. It is offered as a reading to be tested, not as established exposition.
Mark 14:36 · Romans 8:15 · Galatians 4:4-6
⚙ Pilate names Jesus exactly once in this unit, and the name he uses is a title: τὸν Βασιλέα τῶν Ἰουδαίων, the King of the Jews. The commentators divide on his motive. Gill says he either spoke it in derision of Christ and of them, or to shame them out of putting their own king to death; Benson thinks the repetition politically senseless, enraging the very men he sought to appease, and elsewhere wonders whether Pilate meant to ridicule the whole idea of a Messiah. Matthew Henry reads the transaction as forfeiture, that by delivering up the King they in effect delivered up the kingdom of God. The irony belongs to Mark, not to us: the one true statement made about the prisoner in this hearing is made by a pagan who does not believe it, and it will be nailed above his head.
Mark 15:26 · John 19:19-22 · Zechariah 9:9
⚙ Mark says the cause was φθόνος, and Cambridge names its object precisely: the influence He had gained and the favour He had won throughout the land. Envy is the sin that concedes the worth of what it destroys, which makes it a strange and backhanded testimony. Scripture tells the story before: his brothers were jealous of him, but his father kept the matter in mind (Genesis 37:11), and the son sent last was killed by the tenants who said the inheritance would be theirs, a parable Mark himself had already recorded (Mark 12:6-8). Wrath is cruel and anger overwhelming, but who can stand before jealousy (Proverbs 27:4). The line from Joseph through the wicked tenants to the chief priests of 15:10 is a pattern of the envied son, and it is the tradition's reading as much as ours.
Genesis 37:11 · Proverbs 27:4 · Mark 12:6-8 · Acts 7:9
The biblical text is the Berean Standard Bible (BSB), public domain (CC0). Hebrew/Greek text, transliteration, morphology and Strong’s are transcribed from the Berean interlinear (CC0) + Strong’s lexicons (PD); the literal renderings, divergence notes, word notes and all synthesis are this tool’s own work (⚙) — fallible; verify them.
Named voices, quoted verbatim from public-domain works:
Text. One substantive variant governs verse 8. Pulpit argues that ἀναβὰς (went up) is to be preferred over the older ἀναβοήσας (crying aloud), naming the Sinaitic, Vatican and Cambridge manuscripts, the Old Italic and the Gothic, and noting that the Ethiopic combines the two. Gill and Geneva, working from the received text, read "crying aloud," and Benson prints the older Greek in full. The Berean interlinear used in this brief reads ἀναβὰς, so our literal rendering follows Pulpit; the reader should know the older English is not an error but a different text.
Text, second. In verse 7 the number of "had committed murder" is disputed. Gill records that the Vulgate reads the singular, of Barabbas himself, while the Syriac, Arabic and Persic read the plural of his companions, and the Ethiopic renders it "he was bound with seditious persons and murderers." The interlinear gives the plural πεποιήκεισαν with οἵτινες, which is the reading our notes follow.
Sources. The commentary in this unit is thin in places by no fault of the editors. Barnes refers the whole chapter to his notes on Matthew 27 and says nothing else; Poole refers to Mark 15:2; JFB supplies only a chapter heading and a cross-reference to John 18:28. Matthew Henry's note covers Mark 15:1-14 as a single block, so his sentences are not comments on any one verse. Ellicott, Cambridge, Pulpit, Gill and Benson carry the verse-level work here, and our voices are drawn accordingly.
Corrupted source text. The digitized Cambridge note at verse 7 has been damaged by an automatic scripture-linker that replaced the word "Son" with "Song of Solomon" inside the two derivations of the name. We quote it verbatim as the rules require and flag the corruption in an editorial note rather than silently repairing it.
What is uncertain. The origin and scope of the release custom are unknown: Cambridge says so explicitly and doubts it was limited to Passover, Benson and Gill hold that it was a Passover practice, and there is no external documentation in the brief that settles it. Pilate's motive in verse 9 is likewise unsettled among the sources: Cambridge and Pulpit read genuine intent to free Jesus, Benson reads cowardice and folly, Gill leaves derision open. The Greek supports Ellicott's cautious "half hope" and no more.
What is ours. Every thread badge above records its own basis, and the two cross-Testament threads (Leviticus 16 and Proverbs 17:15 with Isaiah 5:23) carry no computed lexeme basis at all, because Greek and Hebrew cannot share Strong's numbers. They are figural and thematic readings offered under the ⚙ mark. The Day of Atonement typology and the Bar-Abbas name reading are not drawn by any commentator in this brief, and we say so where they appear.
✦ = human, public-domain source, quoted and named. ⚙ = machine synthesis, to be verified. Flagged cross-references are left visible on purpose — the verifier working in the open. “Search the Scriptures daily, whether those things were so.” (Acts 17:11)