τὸ δὲ Ἁγὰρ Σινᾶ ὄρος ἐστὶν ἐν τῇ Ἀραβίᾳ, συστοιχεῖ δὲ τῇ νῦν Ἰερουσαλήμ, δουλεύει γὰρ μετὰ τῶν τέκνων αὐτῆς·
to de Hagar Sina oros estin en tē Arabia, systoichei de tē nyn Ierousalēm, douleuei gar meta tōn teknōn autēs;
το δε αγαρ σινα ορος εστιν εν τη αραβια, συστοιχει δε τη νυν ιερουσαλημ, δουλευει γαρ μετα των τεκνων αυτης·Parsing · 20 words
| τὸ | ὁ | art nom/acc neut sg | the [word/term] — Neuter article with the indeclinable proper noun Ἁγάρ: metalinguistic, naming the word itself. |
| δὲ | δέ | conj (postpositive) | now / and — Continuative, carrying the exposition forward from v. 24. |
| Ἁγὰρ | Ἁγάρ | indecl proper noun | Hagar |
| Σινᾶ | Σινᾶ | indecl proper noun | Sinai — Indeclinable transliteration of סִינַי; here attributive to ὄρος. |
| ὄρος | ὄρος | noun nom neut sg | mountain — Predicate nominative. |
| ἐστὶν | εἰμί | pres act ind 3sg | is — Equative in an allegorical identification, not a geographical claim. |
| ἐν | ἐν | prep + dat | in |
| τῇ | ὁ | art dat fem sg | the |
| Ἀραβίᾳ | Ἀραβία | noun dat fem sg | Arabia |
| συστοιχεῖ | συστοιχέω | pres act ind 3sg | stands in the same column — NT hapax; military/tabular metaphor. |
| δὲ | δέ | conj (postpositive) | and |
| τῇ | ὁ | art dat fem sg | the |
| νῦν | νῦν | adv (attributive) | present / now-existing — Adverb in attributive position between article and noun. |
| Ἰερουσαλήμ | Ἰερουσαλήμ | indecl proper noun, dat by construction | Jerusalem |
| δουλεύει | δουλεύω | pres act ind 3sg | is in slavery / serves as a slave |
| γὰρ | γάρ | conj (postpositive) | for — Grounds the alignment just asserted. |
| μετὰ | μετά | prep + gen | with / together with |
| τῶν | ὁ | art gen neut pl | the |
| τέκνων | τέκνον | noun gen neut pl | children — Relational, not juridical; per standing glossary. |
| αὐτῆς | αὐτός | pron gen fem 3sg | her — Feminine: refers to Jerusalem (or Hagar-as-mother), not to neuter ὄρος. |
Now the [word] ‘Hagar’ is Mount Sinai in Arabia; and it stands in the same column with the present Jerusalem, for she serves as a slave together with her children.
Now this ‘Hagar’ is Mount Sinai in Arabia, and she lines up in the same column as the present Jerusalem — for she is in slavery along with her children.
But the [term] Ἁγάρ is Mount Sinai in Arabia — and it belongs to the same file as the present-day Jerusalem, for she is in slavery with her children.
Now Hagar is Mount Sinai in Arabia, and she answereth to the Jerusalem which now is, for she is in bondage with her children.
Overview
Paul is midway through the Hagar/Sarah reading he announced in v. 24 as ἀλληγορούμενα. Verse 25 does three things in one breath.
First, it fixes the identification: the term Ἁγάρ is Mount Sinai, located in Arabia — outside the land of promise, in the territory of Hagar’s and Ishmael’s descendants.
Second, it aligns the columns: Hagar/Sinai συστοιχεῖ — stands in the same file, the same vertical column of a table — with τῇ νῦν Ἰερουσαλήμ, the Jerusalem that now is. This is the sentence’s shock. Paul places the actual, present, temple-standing city on the slave side of the ledger.
Third, it supplies the ground: δουλεύει γὰρ μετὰ τῶν τέκνων αὐτῆς — she is in slavery along with her children. The verb is the same root as δεδουλωμένοι in v. 3 and δουλεύειν in v. 9. The Galatians are being told that adopting Torah-observance moves them not forward into maturity but back into the very column they were rescued from.
The verse also carries the chapter’s densest textual problem: the manuscripts divide over whether Paul wrote Ἁγάρ at all in the first clause, and over the syntax of what follows.
The text of the first clause
This is one of the genuinely difficult variant units in Paul, and the printed critical text is not unanimous.
The principal readings:
- τὸ δὲ Ἁγὰρ Σινᾶ ὄρος ἐστὶν ἐν τῇ Ἀραβίᾳ — with neuter article, name, and no καί. / text. Supported by (with variation), A, , , and much of the tradition.
- τὸ γὰρ Σινᾶ ὄρος ἐστὶν ἐν τῇ Ἀραβίᾳ — omitting Ἁγάρ altogether. Read by א, , , , and part of the Latin and Coptic evidence.
- τὸ δὲ Ἁγὰρ Σινᾶ ὄρος ἐστὶν ἐν τῇ Ἀραβίᾳ with καί inserted before συστοιχεῖ, and other minor rearrangements in the later Byzantine stream.
The internal arguments run in both directions, which is why the editors mark the passage with a low confidence rating.
For omitting Ἁγάρ: the shorter reading is harder in one sense — a bare geographical note (‘for Sinai is a mountain in Arabia’) is an oddly flat parenthesis, and a scribe might well add the name to restore the allegorical thread lost by the neuter article. It also removes the puzzle of a neuter article with a feminine name.
For retaining Ἁγάρ: without it the sentence has no subject for συστοιχεῖ that connects back to the woman of v. 24, and the whole point of the verse — the identification, not the geography — evaporates. The awkward neuter τό is precisely the kind of thing scribes remove, and its very awkwardness argues for originality.
The external evidence is split between the two strongest text-types, and neither internal criterion is decisive. Metzger’s committee retained Ἁγάρ but with an explicit note of doubt; Lightfoot, Burton, and a significant minority of modern commentators prefer the omission and read the clause as a plain geographical parenthesis. The rendering above adopts the printed text, but the reader should know that a defensible Galatians 4:25 exists in which Hagar’s name does not appear in this clause at all.
A further ancient rumor, reported by Chrysostom and some later Greek writers, holds that Ἁγάρ was the local Arabic name for Sinai. No first-century evidence supports this; it looks like an explanation generated by the text rather than data behind it, and it should not be used as an argument for the reading.
τὸ … Ἁγάρ
The neuter article τό before the feminine indeclinable Ἁγάρ is the syntactic hinge of the verse. Greek uses the neuter article to quote a word as a word — to name the linguistic item rather than the person. Compare the common construction where τό introduces a cited term or clause.
So the sense is: ‘the term Hagar denotes Mount Sinai in Arabia.’ Paul is doing lexicography inside his allegory. He is not saying that a woman is a mountain; he is saying that within the figure he is expounding, this name has this referent.
The literal tier keeps the metalinguistic force visible with a bracketed supplement (‘the [word] Hagar’); smooth registers may use quotation marks, which do the same work in English typography.
A few interpreters take τό as agreeing proleptically with the neuter ὄρος rather than as metalinguistic. This is grammatically possible but leaves the word order strained and does not explain why Paul would front the article before the name.
συστοιχέω
συστοιχεῖ is a NT hapax and the most vivid word in the verse. It is built on στοῖχος, a row or file, and belongs to the same family as στοιχεῖα — the very word Paul used in vv. 3 and 9 for the elemental powers under which the Galatians were enslaved. That echo is almost certainly deliberate and should not be lost in a paraphrase.
The lexical background is twofold:
- Military: soldiers drawn up in ranks and files. To συστοιχεῖν with someone is to stand in the same file — behind or in front of them, in the same vertical line.
- Tabular/philosophical: the Pythagorean συστοιχία, the paired columns of opposites (limit/unlimited, light/dark, male/female). Aristotle uses the term for such a column of correlated terms.
Either background yields the same sense here, and Paul’s argument is itself a two-column table: Hagar / Sarah, slave / free, flesh / promise, Sinai / the Jerusalem above. συστοιχεῖ says that Hagar-Sinai and present Jerusalem are entries in the same column.
English has no single verb for this. ‘Corresponds to’ is the standard lexical gloss but has bleached out to mere similarity. ‘Belongs to the same column’ or ‘lines up with’ preserves the image, which per the live-metaphor protocol must remain visible. Avoid ‘answers to,’ which in modern English suggests accountability rather than alignment.
Built onbdag · bdf · muraoka-lxx
ἐν τῇ Ἀραβίᾳ
‘Arabia’ in the first century AD is not the modern peninsula but the broad region east and south of the Jordan, including Nabataea — the territory Paul himself visited (Gal 1:17). Sinai’s exact location is disputed in antiquity as it is now, and Paul is not adjudicating it.
The point is not cartographic but genealogical. Arabia is the territory associated with Ishmael’s line (Gen 25:12-18) and with the Hagarites. Placing Sinai there sets the mountain of the covenant outside the land of promise and inside the sphere of the slave woman’s descendants. The geography is doing theological work: the covenant given at Sinai is, on Paul’s reading of this figure, extraterritorial to the inheritance.
The extra force is that Paul is not disparaging Torah as such — Gal 3:21 has already denied that. He is disputing the claim that Torah-observance is the route into the inheritance.
τῇ νῦν Ἰερουσαλήμ
νῦν is an adverb standing in attributive position between article and noun, functioning adjectivally: ‘the now-Jerusalem,’ the Jerusalem that presently exists. The construction is standard Koine and is answered directly in v. 26 by ἡ ἄνω Ἰερουσαλήμ, ‘the Jerusalem above,’ where ἄνω is likewise an attributive adverb.
The pairing is spatial-temporal against spatial: now against above. Paul does not say ‘the earthly Jerusalem’ (which would be the obvious antonym to ‘above’) nor ‘the future Jerusalem’ (which would be the obvious antonym to ‘now’). The asymmetry is his, and English should reproduce it rather than tidy it — ‘the present Jerusalem’ and ‘the Jerusalem above.’
Paraphrasing to ‘present-day Judaism’ imports a category Paul does not name and flattens a concrete city into an abstraction. The referent is the city, and through the city the covenant order centered there.
Note also the spelling: Paul uses the indeclinable Semitic form Ἰερουσαλήμ here and in v. 26, not the hellenized Ἱεροσόλυμα he uses in Gal 1:17-18 and 2:1. Whether the choice is deliberate — the sacral form for the theologically weighted argument — or simply variation is not decidable, but it is worth recording.
δουλεύω
δουλεύει — the third-person singular subject is feminine, so it is Jerusalem (or, on the other construal, Hagar as the figure standing for her) who is in slavery. It cannot be ὄρος, which is neuter, and the following αὐτῆς confirms the feminine referent.
The standing glossary decision holds: δουλεύω is ‘serve as a slave,’ and the δοῦλ- root is theologically load-bearing here as it is in Philippians. This verse sits inside a chain:
- v. 3: δεδουλωμένοι — we were enslaved under the elements
- v. 7: οὐκέτι εἶ δοῦλος — you are no longer a slave
- v. 8: ἐδουλεύσατε — you served as slaves to non-gods
- v. 9: δουλεύειν θέλετε — you want to be enslaved all over again
- v. 24: εἰς δουλείαν γεννῶσα — bearing children into slavery
- v. 25: δουλεύει
English must use one root across the chain or the argument becomes invisible. ‘Is in bondage’ is acceptable in liturgical register; ‘is in slavery’ is the modern equivalent. ‘Serves’ alone softens an ancient legal institution the argument depends on, contrary to §5 and the corpus policy.
The present tense is durative and current: this is the state of affairs as Paul writes, not a past condition.
Built onbdag
The scope of γάρ
δουλεύει γὰρ μετὰ τῶν τέκνων αὐτῆς grounds what precedes. The question is which preceding clause.
The natural reading takes it as grounding συστοιχεῖ: Hagar-Sinai lines up with the present Jerusalem because that Jerusalem is in slavery, as Hagar was. The shared predicate of slavery is what puts them in one column. This reading gives the γάρ real argumentative work and matches the two-column logic of the whole passage.
On either construal the theological outcome is the same, but the connective must be rendered — epistolary argument structure is load-bearing — and ‘for’ is the correct English.
μετὰ τῶν τέκνων αὐτῆς extends the verdict from the city to those who belong to it. In the Genesis figure Hagar’s slavery was shared by Ishmael; in Paul’s application the present Jerusalem’s slavery is shared by those who look to her as mother. Verse 26 supplies the counterpart: ἥτις ἐστὶν μήτηρ ἡμῶν.
How the Fathers read it
The early commentators were uniformly nervous about ἀλληγορούμενα in v. 24 and worked hard to show that Paul was not dissolving the history.
Chrysostom insists that Paul uses ἀλληγορία loosely, for what is really τύπος — the events happened, and their pattern points forward. On v. 25 he passes along the report that Sinai is called Hagar in the local tongue, and takes συστοιχεῖ to mean that Sinai ‘borders on’ or ‘is contiguous with’ the present Jerusalem — a spatial reading of the file-metaphor that is lexically defensible but probably too literal.
Jerome, in his commentary on Galatians, treats the geography seriously and connects Arabia to the descendants of Ishmael, which is close to the reading given above.
Augustine presses the two-covenant structure hard in his exposition, and it is from this passage that the whole medieval framework of the two cities eventually draws breath. That is reception, not exegesis of v. 25, and is recorded here as reception only.
What the patristic reading contributes is a warning that the modern reader should heed in the other direction: none of them took Paul to be saying that the Sinai covenant was evil, or that the Jewish people were rejected. The verse is an argument about where a Gentile convert should look for the inheritance, addressed to Galatians who were being told to look to Sinai.
Reading responsibly
Per §5, a note is owed here. This verse has a long and ugly reception history in which ‘the present Jerusalem is in slavery with her children’ was read as a verdict on the Jewish people as such.
The text does not support that. Paul’s target throughout Galatians is a specific proposal being made to Gentile converts — that they must come under Torah to belong to Abraham. His own standing as an Israelite is not in question (Gal 2:15), and Romans 9-11 makes the continuing standing of Israel explicit. Verse 25 is a move in an intra-Jewish argument about how Gentiles enter the promise, made by a Jew, using Jewish exegetical technique on a Jewish text.
The translation’s job is to render what is there — including the hardness of putting the actual city of Jerusalem in the slave column — without either softening it or letting it be enlarged into something the Greek does not say.
Nestle-Aland, Novum Testamentum Graece. 28th rev. ed. Stuttgart, 2012.
Metzger, B. M. A Textual Commentary on the Greek New Testament. 2nd ed. Stuttgart, 1994.
Bauer, W., F. W. Danker, W. F. Arndt, F. W. Gingrich. A Greek–English Lexicon of the New Testament and Other Early Christian Literature. 3rd ed. Chicago, 2000.
Muraoka, T. A Greek–English Lexicon of the Septuagint. Leuven, 2009.
Blass, F., A. Debrunner, R. W. Funk. A Greek Grammar of the New Testament. Chicago, 1961.