ii. 2 we shall find this MS alone (with one important Latin father)

retaining the correct text; in the latter case amidst a great complication of various readings. And when again, as in iv. 8, we find B for once on the side of a reading which might otherwise be suspected as a harmonistic change, this support alone will weigh heavily in its favour. Other cases in which B (with more or less support) preserves the correct reading against the mass of authorities are ii. 2 πᾶν πλοῦτος, ii. 7 τῇ πίστει, ii. 13 τοῖς παραπτώμασιν (omitting ἐν), v. 12 σταθῆτε, together with several instances which will appear in the course of the following investigation. On the other hand its value must not be overestimated. [Sidenote: False readings in B.]Thus in iv. 3 τὸ μυστήριον τοῦ Χριστοῦ δι’ ὃ καὶ δέδεμαι[536] there can be little doubt that the great majority of ancient authorities correctly read δι’ ὅ, though B F G have δι’ ὅν: but the variation is easily explained. A single stroke, whether accidental or deliberate, alone would be necessary to turn the neuter into a masculine and make the relative agree with the substantive nearest to it in position. Again in ii. 10 ὅς ἐστιν ἡ κεφαλή, the reading of B which substitutes ὅ for ὅς is plainly wrong, though supported in this instance by D F G 47^*, by the Latin text d, and by Hilary in one passage (_de Trin._ ix. 8, II. p. 263), though elsewhere (ib. i. 13, I. p. 10) he reads ὅ. But here again we have only an instance of a very common interchange. Whether for grammatical reasons or from diplomatic confusion or from some other cause, five other instances of this interchange occur in this short epistle alone; i. 15 ὅ for ὅς F G; i. 18 ὅ for ὅς F G; i. 24 ὅς for ὅ C D^* etc.; i. 27 ὅς for ὅ א C D K L etc.; iii. 14 ὅς for ὅ א^* D. Such readings again as the omission of καὶ αἰτούμενοι i. 9 by B K, or of δι’ αὐτοῦ in i. 20 by B D^* F G etc., or of ἡ ἐπιστολή in iv. 16 by B alone, need not be considered, since the motive for the omission is obvious, and the authority of B will not carry as great weight as it would in other cases. Similarly the insertion of ἡ in i. 18, ἡ ἀρχή, by B, 47, 67^{**}, b^{scr}, and of καί in ii. 15, καὶ ἐδειγμάτισεν, by B alone, do not appear to deserve consideration, because in both instances these readings would suggest themselves as obvious improvements. In other cases, as in the omission of τῆς before γῆς (i. 20), and of ἑνί in ἐν ἑνί σώματι (iii. 15), the scribe of B has erred as any scribe might err.

Footnote 536:

In this passage B (with some few other authorities) has τοῦ Θεοῦ for τοῦ Χριστοῦ, thus substituting a commoner expression (ii. 2, 1 Cor. iv. 1, Rev. x. 7; comp. 1 Cor. ii. 1, v.l.) for a less common (Ephes. iii. 4).

* * * * *

The various readings in this epistle are more perplexing than perhaps in any portion of St Paul’s Epistles of the same length. The following deserve special consideration.

i. 3 ==τῷ θεῷ πατρί.==

[Sidenote: i. 3 τῷ θεῷ πατρί,]]

On this very unusual collocation I have already remarked in the notes (p. 199). The authorities stand as follows:

(1) τῷ θεῷ πατρί B C^*.

(2) τῷ θεῷ τῷ πατρί D^* F G Chrysostom.

One or other is also the reading of the Old Latin (d, e, g, harl.^{**}), of the Memphitic, the two Syriac (Peshito and Harclean), the Æthiopic, and the Arabic (Erpenius, Bedwell, Leipzig) Versions; and of Augustine (_de Unit. Eccl._ 45, IX. p. 368) and Cassiodorus (II. p. 1351, Migne).

(3) τῷ θεῷ καὶ πατρί א A C^2 D^c K L P and apparently all the other MSS; the Vulgate and Armenian Versions; Euthalius (Tischendorf’s MS), Theodore of Mopsuestia (transl.), Theodoret, the Ambrosian Hilary, and others.

A comparison of these authorities seems to show pretty clearly that τῷ θεῷ πατρί was the original reading. The other two were expedients for getting rid of a very unusual collocation of words. [Sidenote: compared with iii. 17,]The scribes have felt the same difficulty again in iii. 17 εὐχαριστοῦντες τῷ θεῷ πατρὶ δι’ αὐτοῦ, and there again we find καί inserted before πατρί. In this latter instance however the great preponderance of ancient authority is in favour of the unusual form τῷ θεῷ πατρί.

[Sidenote: and i. 12.]

It is worth observing also that in i. 12, where τῷ πατρί has the highest support, there is sufficient authority for τῷ θεῷ πατρί to create a suspicion that there too it may be possibly the correct reading. Thus τῷ θεῷ πατρί is read in א 37, while θεῷ τῷ πατρί stands in F G. One or other must have been the reading of some Old Latin and Vulgate texts (f, g, m, fuld.), of the Peshito Syriac, of the Memphitic (in some texts; for others read τῷ πατρί simply), of the Arabic (Bedwell), of the Armenian (Uscan), and of Origen (II. p. 451, the Latin translator); while several other authorities, Greek and Latin, read τῷ θεῷ καὶ πατρί.

[Sidenote: Unique collocation.]

There is no other instance of this collocation of words, ὁ Θεὸς πατήρ, in the Greek Testament, so far as I remember; and it must be regarded as peculiar to this epistle.

i. 4 ==τὴν ἀγάπην [ἣν ἔχετε]==.

[Sidenote: i. 4 τὴν ἀγάπην [ἣν ἔχετε].]

Here the various readings are;

(1) τὴν ἀγάπην B.

(2) τὴν ἀγάπην ἣν ἔχετε A א C D^* F G P 17, 37, 47; the Old Latin and Vulgate, Memphitic (apparently), and Harclean Syriac Versions; the Ambrosian Hilary, Theodore of Mopsuestia (transl.), and others.

(3) τὴν ἀγάπην τήν. D^c K L; the Peshito Syriac (apparently), and Armenian (apparently) Versions; Chrysostom, Theodoret and others.

If the question were to be decided by external authority alone, we could not hesitate. It is important however to observe that (2) conforms to the parallel passage Philem. 5 ἀκούων σου τὴν ἀγάπην καὶ τὴν πίστιν ἣν ἔχεις, while (3) conforms to the other parallel passage Ephes. i. 15 καὶ [τὴν ἀγάπην] τὴν εἰς πάντας τοὺς ἅγιους. Thus, though ἣν ἔχετε is so highly supported and though it helps out the sense, it is open to suspicion. Still the omission in B may be an instance of that impatience of apparently superfluous words, which sometimes appears in this MS.

i. 7 ==ὑπὲρ ἡμῶν δίακονοϲ==.

[Sidenote: i. 7 ὑπὲρ ἡμῶν.]

Here there is a conflict between MSS and Versions.

(1) ἡμῶν A B א^* D^* F G, 3, 13, 33, 43, 52, 80, 91, 109. This must also have been the reading of the Ambrosian Hilary (though the editors make him write ‘pro _vobis_’), for he explains it ‘qui eis ministravit gratiam Christi _vice Apostoli_.’

(2) ὑμῶν א^c C D^b K L P, 17, 37, 47, and many others; the Vulgate, the Peshito and Harclean Syriac, the Memphitic, Gothic, and Armenian Versions; Chrysostom, Theodore of Mopsuestia (transl.), and Theodoret (in their respective texts, for with the exception of Chrysostom there is nothing decisive in their comments), with others.

The Old Latin is doubtful; d, e having _vobis_ and g _nobis_.

Though the common confusion between these two words even in the best MSS is a caution against speaking with absolute certainty, yet such a combination of the highest authorities as we have here for ἡμῶν does not leave much room for doubt: and considerations of internal criticism point in the same direction. See the note on the passage.

i. 12 ==τῷ ἱκανώϲαντι==.

[Sidenote: i. 12 ἱκανώσαντι.]

Against this, which is the reading of all the other ancient authorities, we have

(2) τῷ καλέσαντι D^* F G, 17, 80, with the Latin authorities d, e, f, g, m, and the Gothic, Armenian, and Æthiopic Versions. It is so read also by the Ambrosian Hilary, by Didymus _de Trin._ iii. 4 (p. 346), and by Vigilius Thapsensis _c. Varim._ i. 50 (p. 409).

(3) τῷ καλέσαντι καὶ ἱκανώσαντι, found in B alone.

Here the confusion between ==τωιικανωϲαντι== and ==τωικαλεϲαντι== would be easy, more especially at a period prior to the earliest existing MSS, when the iota adscript was still written; while at the same time καλέσαντι would suggest itself to scribes as the obvious word in such a connexion. It is a Western reading.

The text of B obviously presents a combination of both readings.

i. 14 ==ἐν ᾧ ἔχομεν==.

[Sidenote: i. 14 ἔχομεν or ἔσχομεν?]

For ἔχομεν B, the Memphitic Version, and the Arabic (Bedwell, Leipzig), read ἔσχομεν. This is possibly the correct reading. In the parallel passage, Ephes. i. 7, several authorities (א^* D^*, the Memphitic and Æthiopic Versions, and the translator of Irenæus v. 14. 3) similarly read ἔσχομεν for ἔχομεν. It may be conjectured that ἔσχομεν in these authorities was a harmonistic change in Ephes. i. 7, to conform to the text which they or their predecessors had in Col. i. 14. Tischendorf on Ephes. l.c. says ‘aut utroque loco εχομεν aut εσχομεν Paulum scripsisse puto’; but if any inference can be drawn from the phenomena of the MSS, they point rather to a different tense in the two passages.

i. 22 ==ἀποκατηλλάγητε==.

[Sidenote: i. 22 ἀποκατηλλάγητε.]

This reading is perhaps the highest testimony of all to the great value of B.

The variations are;

(1) ἀποκατηλλάγητε B. This also seems to be the reading of Hilary of Poitiers _In xci Psalm._ 9 (I. p. 270), who transfers the Apostle’s language into the first person, ‘cum aliquando essemus alienati et inimici sensus ejus in factis malis, nunc autem reconciliati sumus corpore carnis ejus.’

(2) ἀποκατηλλάκηται 17.

(3) ἀποκαταλλαγέντες D^* F G, and the Latin authorities d, e, g, m, the Gothic Version, the translator of Irenæus (v. 14. 3), and others.

(4) ἀποκατήλλαξεν, all the other authorities.

Of these (2) is obviously a corruption of (1) from similarity of sound; and (3) is an emendation, though a careless emendation, of (1) for the sake of the grammar. It should have been ἀποκαταλλαγέντας. The reading therefore must lie between ἀποκατηλλάγητε and ἀποκατήλλαξεν. This latter however is probably a grammatical correction to straighten the syntax. In the Memphitic a single letter ⲁϫ for ⲁϥ would make the difference between ἀποκατηλλάγητε and ἀποκατήλλαξεν; but no variation from the latter is recorded.

ii. 2 ==τοῦ θεοῦ, χριϲτοῦ==.

[Sidenote: ii. 2 τοῦ Θεοῦ Χριστοῦ.]

The various readings here are very numerous and at first sight perplexing; but the result of an investigation into their several claims is far from unsatisfactory. The reading which explains all the rest may safely be adopted as the original.

[Sidenote: Original reading.]

(1) ==του θεου χριϲτου==.

This is the reading of B and of Hilary of Poitiers, _de Trin._ ix. 62 (I. p. 306), who quotes the passage _sacramenti Dei Christi in quo_ etc., and wrongly explains it ‘Deus Christus sacramentum est’.

[Sidenote: Variations;]

All the other variations are derived from this, either by explanation or by omission or by amplification.

[Sidenote: (_a_) by interpretation,]

By explanation we get;

(2) ==του θεου ο εϲτιν χριϲτοϲ==,

the reading of D, with the Latin authorities d, e, which have _Dei quod est Christus_. So it is quoted by Vigilius Thapsensis _c. Varim._ i. 20 (p. 380), and in a slightly longer form by Augustine _de Trin._ xiii. 24 (VIII. p. 944) _mysterium Dei quod est Christus Jesus_.

(3) ==του θεου εν χριϲτω==.

So it is twice quoted by Clement of Alexandria _Strom._ v. 10 (p. 683), _ib._ 12 (p. 694); or

==του θεου του εν χριϲτω==,

the reading of 17.

So the Ambrosian Hilary (both text and commentary) has _Dei in Christo_. And the Armenian has the same lengthened out, _Dei in Christo Jesu_ (Zohrab) or _Dei patris in Christo Jesu_ (Uscan).

(4) _Domini quod de Christo_

is the Æthiopic rendering. Whether this represents another various reading in the Greek or whether the paraphrase is the translator’s own, it is impossible to say.

[Sidenote: (_b_) by omission,]

The two following variations strive to overcome the difficulty by omission;

(5) ==του θεου==,

the reading of D by a second hand, of P, 37, 67^{**}, 71, 80, 116.

(6) ==του χριϲτου==,

the reading of Euthalius in Tischendorf’s MS; but Tischendorf adds the caution ‘sed non satis apparet’.

[Sidenote: (_c_) by amplification;]

All the remaining readings are attempts to remedy the test by amplification. They fall into two classes; those which insert πατρός so as to make Χριστοῦ dependent on it, (7), (8), and those which separate Θεοῦ from Χριστοῦ by the interposition of a καί, (9), (10), (11).

[Sidenote: (i) by inserting πατρός to govern Χριστοῦ;]

(7) ==του θεου πατροϲ χριϲτου==,

the reading of א (by the first hand). Tischendorf also adds b^{scr*} and o^{scr}; but I read Scrivener’s collations differently (_Cod. Aug._ p. 506): or

==του θεου πατροϲ του χριϲτου==,

the reading of A C, 4.

One or other is the reading of the Thebaic Version (given by Griesbach) and of the Arabic (Leipz.).

A lengthened form of the same, _Dei patris Christi Jesu_, appears in the oldest MSS of the Vulgate, am. fuld. f: and the same is also the reading of the Memphitic (Boetticher).

(8) ==του θεου και πατροϲ του χριϲτου==.

So א (the third hand), b^{scr*}, o^{scr}, and a corrector in the Harclean Syriac.

[Sidenote: (ii) by separating Θεοῦ from Χριστοῦ by a conjunction.]

(9) ==του θεου και χριϲτου==,

the simplest form of the other class of emendations by amplification. It is found in Cyril _Thes._ p. 287.

(10) ==του θεου πατροϲ και του χριϲτου==.

So 47, 73, the Peshito Syriac (ed. princeps and Schaaf). And so it stands in the commentators Chrysostom (but with various readings) and Theodore of Mopsuestia (_Spicil. Solesm._ I. p. 131 _Dei patris et Christi_, but in Rab. Maur. _Op._ VI. p. 521 _Dei patris Christi Jesu_).

Pelagius has _Dei patris et Christi Jesu_>, and so the Memphitic (Wilkins).

[Sidenote: The common text the latest development.]

(11) ==του θεου και πατροϲ και του χριϲτου==.

This, which may be regarded as the latest development, is the reading of the received text. It is found in D (third hand) K L, and in the great majority of cursives; in the text of the Harclean Syriac, and in Theodoret and others.

Besides these readings some copies of the Vulgate exhibit other variations; e.g. demid. _Dei patris et Domini nostri Christi Jesu_, tolet. _Dei Christi Jesu patris et Domini_.

It is not necessary to add any remarks. The justification of τοῦ Θεοῦ Χριστοῦ as the original reading will have appeared in the variations to which it has given rise. The passage is altogether an instructive lesson in textual criticism.

ii. 16 ==ἐν βρώϲει καὶ ἐν πόϲει.==

[Sidenote: ii. 16 καί or ἤ?]

In this reading B stands alone among the MSS; but it is supported by the Peshito Syriac and Memphitic Versions, by Tertullian (_adv. Marc._ v. 19), and by Origen (_in Ioann._ x. § 11, IV. p. 174). The testimony of Tertullian however is invalidated by the fact that he uses _et_ as the connecting particle throughout the passage; and the Peshito Syriac also has ‘and’ for ἤ in the two last clauses, though not in the second.

The rest have ἐν βρώσει ἢ ἐν πόσει. This may be explained as a very obvious, though not very intelligent, alteration of scribes to conform to the disjunctive particles in the context, ἢ ἐν μέρει ἑορτῆς ἢ νεομηνίας ἢ σαββάτων.

In this same context it is probable that B retains the right form νεομηνίας (supported here by F G and others) as against the Attic νουμηνίας. In the same way in iii. 25 κομίσεται and iv. 9 γνωρίσουσιν B (with some others) has resisted the tendency to Attic forms.

ii. 18 ==ἃ ἕορακεν==.

[Sidenote: ii. 18, the omission of the negative.]

That this is the oldest reading which the existing texts exhibit, will appear from the following comparison of authorities.

(1) ἃ ἑώρακεν ἑώρακεν) A B א^* D^*, 17^*, 28, 67^{**}; the Old Latin authorities d, e, m; the Memphitic, Æthiopic, and Arabic (Leipz.) Versions; Tertull. _c. Marc._ v. 19 (‘ex visionibus angelicis’; and apparently Marcion himself also); Origen (_c. Cels._ v. 8, I. p. 583, though the negative is here inserted by De la Rue, and _in Cant._ ii, III. p. 63, _in his quæ videt_); Lucifer (_De non conv. c. hær._ p. 782 Migne); the Ambrosian Hilary (_ad loc._ explaining it ‘Inflantur motum pervidentes stellarum, quas angelos vocat’). So too the unknown author of _Quæst. ex N. T._ ii. 62 in August. _Op._ III. Appx. p. 156. Jerome (_Epist. cxxi ad Alg._ § 10, I. p. 880) mentions both readings (with and without the negative) as found in the Greek text: and Augustine (_Epist._ 149, II. p. 514), while giving the preference to _quæ non vidit_>, says that some MSS have _quæ vidit_>.

(2) ἃ μὴ ἑώρακεν (ἑόρακεν) א^c C D^{bc} K L P, and the great majority of cursives;

(3) ἃ οὐκ ἑώρακεν F G.

The negative is also read in g; in the Vulgate, the Gothic, both the Syriac, and the Armenian Versions; in the translator of Origen _In Rom._ ix. § 42 (IV. p. 665), in Ambrose _In Psalm. cxviii Exp._ xx (I. p. 1222), and in the commentators Pelagius, Chrysostom, Theodore of Mopsuestia (_Spic. Solesm._ I. p. 132 ‘quæ nec sciunt’), Theodoret, and others.

From a review of these authorities we infer that the insertion of the negative was a later correction, and that ἃ ἕωρακεν (or ἕορακεν) represents the prior reading. In my note I have expressed my suspicion that ἃ ἕωρακεν (or ἕορακεν) is itself corrupt, and that the original reading is lost.

[Sidenote: The form ἕορακεν.]

The unusual form ἕορακεν is found in א B^* C D P, and is therefore to be preferred to ἕωρακεν.

ii. 23 ==[καὶ] ἀφειδίᾳ ϲώματοϲ==.

[Sidenote: ii. 23. Is καί to be omitted?]

Here καί is found in all the Greek copies except B, but is omitted in these Latin authorities, m, the translator of Origen (_In Rom._ ix. § 42, IV. p. 665), Hilary of Poitiers (_Tract. in xiv Ps._ § 7, p. 73), the Ambrosian Hilary, Ambrose (_de Noe_ 25, p. 267), and Paulinus (_Epist._ 50, p. 292 sq.). We have more than once found B and Hilary alone in supporting the correct reading (i. 22, ii. 2); and this fact gives weight to their joint authority here. The omission also seems to explain the impossible reading of d, e, which have _in religione et humilitate sensus et vexationem corporis_, where for _et vexationem_, we should probably read _ad vexationem_, as in the Ambrosian Hilary. There was every temptation for a scribe to insert the καί so as to make ἀφειδίᾳ range with the other datives: while on the other hand a finer appreciation of the bearing of the passage suggests that St Paul would have dissociated it, so as to give it a special prominence.

A similar instance occurs in iii. 12 ὡς ἐκλεκτοὶ τοῦ Θεοῦ, ἅγιοι καὶ ἠγαπημένοι, where B omits the καί with 17 and the Thebaic Version[537]. In 219 καὶ ἅγιοι is read for ἅγιοι καί. The great gain in force leads to the suspicion that this omission may be correct, notwithstanding the enormous preponderance of authority on the other side.

Footnote 537:

It is true that in the text (_Spicil. Solesm._ I. p. 123, Rab. Maur. Op. VII. p. 539, Migne) he is credited with the later Latin reading _ut cognoscat quæ circa vos sunt_, but his comment implies the other; ‘Quoniam omnia vobis nota faciet Tychicus illa quæ erga me sunt, propterea a me directus est cum Onesimo fratre qui a vobis venerat, ut nota vobis faciant quæ erga nos sunt [= γνῶτε τὰ περὶ ἡμῶν] et oblectent vos per suum adventum [= καὶ παρακαλέσῃ τὰς καρδίας ὑμῶν], omnia quæ hic aguntur manifesta facientes vobis.’ See _Spicil. Solesm._ l.c.; the comment is mutilated in Rab. Maur. _Op._ l.c.

iv. 8. ==γνῶτε τὰ περὶ ἡμῶν==.

[Sidenote: iv. 8 γνῶτε τὰ περὶ ἡμῶν.]

Of the various readings of this passage I have already spoken (p. 29 sq., note 1, p. 301).

The authorities are as follows:

(1) γνῶτε τὰ περὶ ἡμῶν A B D^* F G P, 10, 17, 33, 35, 37, 44, 47, 71, 111, 116, 137; d, e, g; the Armenian and Æthiopic Versions; Theodore of Mopsuestia[537], Theodoret[538], Jerome (on Ephes. vi. 21 sq., VII. p. 682), and Euthalius (Tischendorf’s MS). This is also the reading of א^*, except that it has ὑμῶν for ἡμῶν.

(2) γνῷ τὰ περὶ ὑμῶν א^c C D^{bc} K L and the majority of cursives; the Memphitic, Gothic, Vulgate, and both Syriac Versions; the Ambrosian Hilary, Jerome (on Philem. I, VII. p. 748), Chrysostom (expressly), and others.

Footnote 538:

In the text; but in the commentary he is made to write ἵνα γνῷ γάρ, φησί, τὰ περὶ ἡμῶν, an impossible reading.

[Sidenote: The various readings accounted for.]

The internal evidence is considered in the note on the passage, and found to accord with the vast preponderance of external authority in favour of γνῶτε τὰ περὶ ἡμῶν. The reading of א by the first hand exhibits a transitional stage. It would appear as though the transcriber intended it to be read γνῷ τε τὰ περὶ ὑμῶν. At all events this is the reading of III and of Io. Damasc. _Op._ II. p. 214. The variation γνῷ τὰ περὶ ὑμῶν is thus easily explained. (1) ἡμῶν would be accidentally substituted for ὑμῶν; (2) γνῶτε would then be read γνῷ τε; (3) the awkward and superfluous τε would be omitted. In illustration of the tendency to conform the persons of the two verbs γνῷ, παρακαλέσῃ, (see p. 301) it may be mentioned that 17 reads γνῶτε, παρακαλέσητε, both here and in Ephes. vi. 22.

iv. 15. ==κατ’ οἶκον αὐτῶν==.

[Sidenote: iv. 15 αὐτῶν.]

The readings here are:

(1) αὐτῶν א A C P, 5, 9, 17, 23, 34, 39, 47, 73; together with the Memphitic Version, the Arabic (Leipz.), and Euthalius (Tischendorf’s MS). The Memphitic Version is commonly but wrongly quoted in favour of αὐτοῦ, owing to a mistranslation of Wilkins. But both Wilkins and Boetticher give without any various reading ΠΟΥΗΙ, i.e. οἶκον αὐτῶν. This seems also to be the reading of Theodore of Mopsuestia (_Spic. Solesm._ I. p. 133) _quæ in domo eorum est ecclesia_; though in Rab. Maur. _Op._ VI. p. 540 his text runs _quæ in domo ejus est ecclesiam_, and he is made to say _Nympham cum omnibus suis qui in domo ejus sunt_.

(2) αὐτῆς B 67^{**}.

(3) αὐτοῦ D F G K L and the great majority of cursives; and so the Gothic Version, Chrysostom, and Theodoret (the latter distinctly).

[Sidenote: _Nymphas_ or _Nympha_?]]

The singular, whether αὐτοῦ or αὐτῆς, is the reading of the old Latin and Vulgate, which have _ejus_, and of the Armenian. The pronoun is also singular in the Peshito and Harclean Syriac. In this language the same consonants express masculine and feminine alike, the difference lying in the pointing and vocalisation. And here the copies are inconsistent with themselves. [Sidenote: The Syriac versions.]In the Peshito (both the editio princeps and Schaaf) the proper name is vocalised as a feminine _Numphē_ (= Νύμφη), and yet ܒܒܝܬܗ is treated as having a masculine affix κατ’ οἶκον αὐτοῦ. In the text of the Harclean ܕܝܠܗܿ is pointed thus, as a feminine αὐτῆς; while the margin gives the alternative reading ܕܝܠܗ (without the point) = αὐτοῦ. The name itself is written Nympha, which according to the transliteration of this version might stand either for a masculine (as _Barnaba_, _Luka_, in the context, for Βαρναβας, Λουκᾶς) or for a feminine (since _Demas_, _Epaphras_, are written with an _s_)[539]. [Sidenote: The Latin authorities.]The Latin _ejus_ leaving the gender undetermined, the Latin commentators were free to take either Nymphas or Nympha; and, as Nympha was common Latin form of Νύμφη, they would naturally adopt the female name. So the commentator Hilary distinctly.

Footnote 539:

More probably the latter. In Rom. xvi the terminations -α and ᾶς for the feminine and masculine names respectively are carefully reproduced in the Harclean Version. In ver. 15 indeed we have _Julias_, but the translator doubtless considered the name to be a contraction for _Julianus_. The proper Syriac termination _-a_ seems only to be employed for the Greek -ας in very familiar names such as _Barnaba_, _Luka_.

It should be added that the word is accentuated as a masculine νυμφᾶν in D^c L P, and as a feminine νύμφαν in B^c and Euthalius (Tischendorf’s MS.).

_On the meaning of πλήρωμα._

[Sidenote: The meaning of the verb πληροῦν.]

The verb πληροῦν has two senses. It signifies either (1) ‘To fill’, e.g. Acts ii. 2 ἐπλήρωσεν ὅλον τὸν οἶκον; or (2) ‘To fulfil, complete, perfect, accomplish’, e.g. Matt. xxvi. 56 ἵνα πληρωθῶσιν αἱ γραφαί, Rom. xiii. 8 νόμον πεπλήρωκεν, Acts xii. 25 πληρώσαντες τὴν διακονίαν. The latter sense indeed is derived from the former, but practically it has become separate from it. The word occurs altogether about a hundred times in the New Testament, and for every one instance of the former sense there are at least four of the latter.

[Sidenote: False issue raised respecting πλήρωμα]

[Sidenote: resulting in theological confusion]

In the investigations which have hitherto been made into the signification of the derived substantive πλήρωμα, as it occurs in the New Testament, an almost exclusive prominence has been given to the former meaning of the verb; and much confusion has arisen in consequence. The question has been discussed whether πλήρωμα has an active or a passive sense, whether it describes the filling substance or the filled receptacle: and not unfrequently critics have arrived at the result that different grammatical senses must be attached to it in different passages, even within the limits of the same epistle. Thus it has been maintained that the word has a passive sense ‘id quod impletur’ in Ephes. i. 23 τῇ ἐκκλησίᾳ ἥτις ἐστὶν τὸ σῶμα αὐτοῦ, τὸ πλήρωμα τοῦ τὰ πάντα ἐν πᾶσιν πληρουμένου, and an active sense ‘id quod implet’ in Ephes. iii. 19 ἵνα πληρωθῆτε εἰς πᾶν τὸ πληρώμα τοῦ Θεοῦ. Indeed so long as we see in πληροῦν only the sense ‘to fill’, and refuse to contemplate the sense ‘to complete’, it seems impossible to escape from the difficulties which meet us at every turn, otherwise than by assigning to its derivative πλήρωμα both an active and a passive sense; but the greatest violence is thus done to the connexion of theological ideas.

[Sidenote: and disregard of grammar.]

Moreover the disregard of lexical rules is not less violent[540]. Substantives in -μα, formed from the perfect passive, appear always to have a passive sense. They may denote an abstract notion or a concrete thing; they may signify the action itself regarded as complete, or the product of[Sidenote: Meaning of substantives in -μα.] the action; but in any case they give the _result_ of the agency involved in the corresponding verb. Such for example are ἄγγελμα ‘a message’, ἅμμα ‘a knot’, ἀργύρωμα ‘a silver-made vessel’, βούλευμα ‘a plan’, δικαίωμα ‘a righteous deed’ or ‘an ordinance’, ζήτημα ‘an investigation’, κήρυγμα ‘a proclamation’, κώλυμα ‘a hindrance’, ὁμοίωμα ‘a likeness’, ὅραμα ‘a vision’, στρῶμα ‘a carpet’, σφαίρωμα ‘a round thing’, etc. In many cases the same word will have two meanings, both however passive; it will denote both the completed action and the result or object of the action: e.g. ἅρπαγμα the ‘robbery’ or the ‘booty’, ἀντάλλαγμα the ‘exchange’ or the ‘thing given or taken in exchange’, θήρευμα the ‘hunt’ or the ‘prey’, πάτημα the ‘tread’ or the ‘carpet’, and the like. But in all cases the word is strictly passive; it describes that which might have stood after the active verb, either as the direct object or as the cognate notion. [Sidenote: Apparent exceptions.]The apparent exceptions are only apparent. Sometimes this deceptive appearance is in the word itself. Thus κάλυμμα ‘a veil’ seems to denote ‘that which _covers_’, but it is really derived from another sense and construction of καλύπτειν, not ‘to hide’, but ‘to wrap round’ (e.g. Hom. _Il._ v. 315 πρόσθε δέ οἱ πέπλοιο φαεινοῦ πτύγμ’ ἐκάλυψεν, xxi. 321 τόσσην οἱ ἄσιν καθύπερθε καλύψω), and therefore is strictly passive. Sometimes again we may be led astray by the apparent connexion with the following genitive. Thus in Plut. _Mor._ 78 E δήλωμα τοῦ προκόπτειν the word does not mean, as might appear at first sight, ‘a thing showing’, but ‘a thing shown’, ‘a demonstration given’; nor in 2 Thess. i. 5 ἐνδειγμα τῆς δικαίας κρίσεως must we explain ἕνδειγμα ‘a thing proving’, but ‘a thing proved’, ‘a proof’. And the same is probably the case also with such expressions as συμποσίων ἐρέθισμα (Critias in Athen. xiii. p. 600 D), τόξου ῥῦμα (Æsch. _Pers._ 147), and the like; where the substantives in -μα are no more deprived of their passive sense by the connexion, than they are in ὑπόδημα ποδῶν or στρῶμα κλίνης; though in such instances the license of poetical construction may often lead to a false inference. Analogous to this last class of cases is Eur. _Troad._ 824 Ζηνὸς ἔχεις κυλίκων πλήρωμα καλλίσταν λατρείαν, not ‘the filling’, but ‘the fulness of the cups, the brimming cups, of Zeus.’

Footnote 540:

The meaning of this word πλήρωμα is the subject of a paper _De vocis πλήρωμα vario sensu in N. T._ in Storr’s _Opusc. Acad._ I. p. 144 sq., and of an elaborate note in Fritzsche’s _Rom._ II. p. 469 sq. Storr attempts to show that it always has an active sense ‘id quod implet’ in the New Testament. Fritzsche rightly objects to assigning a persistently active sense to a word which has a directly passive termination: and he himself attributes to it two main senses, ‘id quod impletur’ and ‘id quo res impletur’, the latter being the more common. He apparently considers that he has surmounted the difficulties involved in Storr’s view, for he speaks of this last as a passive sense, though in fact it is nothing more than ‘id quod implet’ expressed in other words. In Rom. xiii. 10 πλήρωμα νόμου he concedes an active sense ‘legis completio’, h. e. ‘observatio’.

[Sidenote: πλήρωμα connected with the second sense of πληροῦν.]]

Now if we confine ourselves to the second of the two senses above ascribed to πληροῦν, it seems possible to explain πλήρωμα in the same way, at all events in all the theological passages of St Paul and St John, without doing any violence to the grammatical form. As πληροῦν is ‘to complete’, so πλήρωμα is ‘that which is completed’, i.e. the complement[541], the full tale, the entire number or quantity, the plenitude, the perfection.

Footnote 541:

The English word complement has two distinct senses. It is either (i) the complete set, the entire quantity or number, which satisfies a given standard or cadre, as e.g. the complement of a regiment; or (ii) the number or quantity which, when added to a preexisting number or quantity, produces completeness; as e.g. the complement of an angle, i.e. the angle by which it falls short of being a complete right angle. In other words, it is either the whole or the part. As a theological term, πλήρωμα corresponds to the first of these two senses; and with this meaning alone the word ‘complement’ will be used in the following dissertation.

[Sidenote: Its uses in Classical writers.]

This indeed is the primary sense to which its commonest usages in classical Greek can be most conveniently referred. Thus it signifies (1) [Sidenote: (1) ‘A ship’s crew.’] ‘A ship’s crew’: e.g. Xen. _Hell._ i. 6. 16 διὰ τὸ ἐκ πολλῶν πληρωμάτων ἐς ὀλίγας (ναῦς) ἐκλελέχθαι τοὺς ἀρίστους ἐρέτας. In this sense, which is very frequent, it is generally explained as having an active force, ‘that which fills the ships’; and this very obvious explanation is recommended by the fact that πληροῦν ναῦν is a recognized expression for ‘manning a ship’, e.g. Xen. _Hell._ i. 6. 24. But πλήρωμα is used not only of the crew which mans a ship, but also of the ship which is manned with a crew; e.g. Polyb. i. 49. 4, 5 τὴν παρουσίαν τῶν πληρωμάτων ... τὰ προσφάτως παραγεγονότα πληρώματα, Lucian _Ver. Hist._ ii. 37, 38, ἀπὸ δύο πληρωμάτων ἐμάχοντο ... πέντε γὰρ εἶχον πληρώματα; and it is difficult to see how the word could be transferred from the crew to the ship as a whole, if the common explanation were correct. Fritzsche (_Rom._ II. p. 469 sq.), to whom I am chiefly indebted for the passages quoted in this paragraph, has boldly given the word two directly opposite senses in the two cases, explaining it in the one ‘ea quibus naves complentur, _h. e._ vel socii navales vel milites classiarii vel utrique’, and in the other ‘id quod completur, _v. c._ navigium’; but this severance of meaning can hardly be maintained. On the other hand, if we suppose that the crew is so called as ‘the complement,’ (i.e. ‘not that which fills the ship,’ but ‘that which is itself full or complete in respect of the ship’), we preserve the passive sense of the word, while at the same time the transference to the fully equipped and manned vessel itself becomes natural. In this sense ‘a complement’ we have the word used again of an army, [Sidenote: (2) ‘Population.’]Aristid. _Or._ I. p. 381 μήτε αὐτάρκεις εσἔσθαι πλήρωμα ἑνὸς οἰκέιου στρατεύματος παρασχέσθαι. (2) It sometimes signifies ‘the population of a city’, Arist. _Pol._ iii. 13 (p. 1284) μὴ μέντοι δυνατοὶ πλήρωμα παρασχέσθαι πόλεως (comp. iv. 4, p. 1291). Clearly the same idea of completeness underlies this meaning of the word, so that here again it signifies ‘the complement’: comp. Dion. Hal. _A. R._ vi. 51 τοῦ δ’ ὀλίγου καὶ οὐκ ἀξιομάχου πληρώματος τὸ πλεῖόν ἐστι δημοτικόν κ.τ.λ., Eur. _Ion_ 663 τῶν φίλων πλήρωμ’ ἀθροίσας ‘the whole body of his friends’. [Sidenote: (3) ‘Total amount.’](3) ‘The entire sum’, Arist. _Vesp._ 660 τούτων πλήρωμα τάλαντ’ ἐγγὺς δισχίλια γίγνεται ἡμῖν, ‘From these sources a total of nearly two thousand talents accrues to us’. [Sidenote: (4) ‘Entire term.’](4) ‘The full term’, Herod. iii. 22 ὀγδώκοντα δ’ ἔτεα ζόης πλήρωμα ἀνδρὶ μακρότατον προκέεσθαι. [Sidenote: (5) ‘Fulfilment.’](5) ‘The perfect attainment’, ‘the full accomplishment’, e.g. Philo _de Abr._ 46 (II. p. 39) πλήρωμα χρηστῶν ἐλπίδων. In short the fundamental meaning of the word generally, though perhaps not universally, is neither ‘the filling material’, nor ‘the vessel filled’; but ‘that which is complete in itself’, or in other words ‘plenitude, fulness, totality, abundance’.

[Sidenote: Use of πλήρωμα in the Gospels. Matt. ix. 16.]

In the Gospels the uses of the word present some difficulty. (1) In Matt. ix. 16 αἴρει γὰρ τὸ πλήρωμα αὐτοῦ ἀπὸ τοῦ ἱματίου καὶ χεῖρον σχίσμα γίνεται, it refers to the ἐπίβλημα ῥάκους ἀγνάφου which has gone before; but πλήρωμα need not therefore be equivalent to ἐπίβλημα so as to mean the patch itself, as is often assumed. The following pronoun αὐτοῦ is most naturally referred to ἐπίβλημα; and if so πλήρωμα describes ‘the completeness’, which results from the patch. The statement is thus thrown into the form of a direct paradox, the very completeness making the garment more imperfect than before. [Sidenote: Mark ii. 21.]In the parallel passage Mark ii. 21 the variations are numerous, but the right reading seems certainly to be αἴρει τὸ πλήρωμα ἀπ’ αὐτοῦ, τὸ καινὸν τοῦ παλαιοῦ κ.τ.λ. The received text omits the preposition before αὐτοῦ, but a glance at the authorities is convincing in favour of its insertion. In this case the construction will be αἴρει τὸ πλήρωμα (nom.) ἀπ’ αὐτοῦ (i.e. τοῦ ἱματίου, which has been mentioned immediately before), τὸ καινὸν (πλήρωμἀ τοῦ παλαιοῦ (ἱματίοὐ; ‘The completeness takes away from the garment, the new _completeness_ of the old _garment_’, where the paradox is put still more emphatically. [Sidenote: Mark vi. 43.] (2) In Mark vi. 43 the right reading is καὶ ἦραν κλασμάτων δώδεκα κοφίνους πληρώματα, i.e. ‘full’ or ‘complete measures’, where the apposition to κοφίνους obviates the temptation to explain πληρώματα as ‘ea quæ implent’. [Sidenote: Mark viii. 20.]On the other hand in Mark viii. 20 πόσων σπυρίδων πληρώματα κλασμάτων ἤρατε; this would be the _prima facie_ explanation; comp. Eccles. iv. 6 ἀγαθόν ἐστι πλήρωμα δρακὸς ἀναπαύσεως ὑπὲρ πληρώματα δύο δρακῶν μόχθου. But it is objectionable to give an active sense to πλήρωμα under any circumstances; and if in such passages the patch itself is meant, it must still be so called, not because it fills the hole, but because it is itself fulness or full measure as regards the defect which needs supplying.

[Sidenote: Usage in St. Paul’s Epistles.]

From the Gospels we pass to the Epistles of St Paul, whose usage bears more directly on our subject. And here the evidence seems all to tend in the same direction. [Sidenote: 1 Cor. x. 26.](1) In 1 Cor. x. 26 τοῦ Κυρίου γὰρ ἡ γῆ καὶ τὸ πλήρωμα αὐτῆς it occurs in a quotation from Ps. xxiv (xxiii). 1. The expressions τὸ πλήρωμα τῆς γῆς, τὸ πλήρωμα τῆς θαλάσσης, occur several times in the LXX (e.g. Ps. xcvi (xcv). 11, Jer.