4) With two double members; for example, Ps. xxxi. 23 :

"I said in my distress,

I am cut off from before thine eyes,

But thou didst hear the voice of my supplication,

When I cried unto thee."

When the members of this rhythmical parallelism are more than double, which is sometimes the case, it approaches very near to prose; it is too loose a form to retain an exuberant matter without passing over into the prosaic style. With good poets, this is rarely the case, but it sometimes occurs; for example, Am. vi. 10; with the later and less correct, it happens more frequently; for example, Mal. i. 6, Zech. xiii. 3, x. 6, Zeph. iii. 8. The length of the members contributes in a special manner to destroy the rhythmical form. But while this form of parallelism brings us to the utmost limits of the province of rhythm, it also settles the question, that the parallelism of members is really a rhythmical form, which there would be room to doubt, if we had nothing but parallelism of thoughts.

The simply rhythmical parallelism holds the most prominent place in the Lamentations of Jeremiah. Here the parallelism of thoughts is to be reckoned almost among the exceptions, and when it does occur, it is, for the most part, the subordinate parallelism of a member by itself; in general, the rhythm alone predominates, and that too with a regularity which is rare among Hebrew poets, producing here a suitable effect, namely, monotony of complaint. The following orders of rhythm may be traced in the Lamentations.* In chapters first and second, the verses consist of three members, the first two of which constitute one parallel, and stand over against the third, as the second parallel. Each member has besides a cæsura, which coincides with the sense and the accent. Still, however, we are sometimes under the necessity of abandoning the accents, because they follow the sense, while the rhythm is independent of the sense. According to the accents, the first parallel is sometimes simple, for example, ch. ii. 6, yet without

*

Comp. Lowth. Prælect., XXII., p. 257, seq.

a valid logical ground. The periods in ch. i. 7, and ch. ii. 19, are distinguished by having four members. It is remarkable that the length of these verses should so greatly exceed those which elsewhere occur in Hebrew poetry. Lowth is of the opinion that these long verses are adapted to lamentation, and it must be acknowledged that they do have a tendency to produce a certain impression of melancholy. Ch. iii. has only verses of one member, without parallelism; yet this one member is rhythmically divided in such a manner as to produce, if not a complete rhythmical parallelism, yet a supplementary clause which conduces to repose. Here again the accents sometimes stand in the way; for example, ch. iii. 3, where n is not enough to form a supplementary clause. Tiphcha, also, sometimes changes place with Zakeph Katon, although the rhythmical cæsura is always the same. Perhaps, however, every three verses are to be considered as a rhythmical whole, as they are connected by having the same initial letters. Ch. v. is of the same structure with ch. iii., except that it has a real short rhythmical parallelism, which, however, the authors of the accents did not consider as complete, and therefore have not separated with Athnach. Ch. iv. has double parallelism, but for the most part simply rhythmical.

We must notice one more exception in Hebrew rhythm. There sometimes occur separate propositions of a single member, almost always introduced with design, since the poet lingers upon the thought; we may conceive it to be accompanied with a long pause; for example, Ps. xxiii. 1, xxv. 1. Here the poet indicates, as it were, the tone and character of the song, and after a pause, again collects himself. Cant. vii. 6 is beautiful:

"How fair, how pleasant art thou, love, in delights!"

where the poet loses himself, as it were, in the contemplation of beauty. In Job x. 22 the voice sinks with two parallel clauses beautifully to repose.

66

"In this peculiar conformation, or parallelism of the sentences," says Lowth, "I apprehend a considerable part of the Hebrew metre to consist; though it is not improbable that some regard was also paid to the numbers and feet. But of this particular we have at present so little information, that it is utterly impos

sible to determine whether it were modulated by the ear alone, or according to any settled or definite rules of prosody.”

"The nervous simplicity and conciseness of the Hebrew muse," says the poet Campbell, "prevent this parallelism from degenerating into monotony. In repeating the same idea in different words, she seems as if displaying a fine opal, that discovers fresh beauty in every new light to which it is turned. Her amplifications of a given thought are like the echoes of a solemn melody, - her repetitions of it like the landscape reflected in the stream; and whilst her questions and responses give a lifelike effect to her compositions, they remind us of the alternate voices in public devotion, to which they were manifestly adapted."

The parallelism affords an important aid in interpretation. For sometimes the meaning of one member of a verse is clear, where that of the other is ambiguous. Thus, the new translation of Ps. xxiv. 4 is confirmed by the parallelism, though it does not depend upon it. In Ps. lv. 15:

"May sudden death seize upon them!

May they go down to the underworld alive!"

the second line is no doubt intended to be synonymous with the first, and is completely explained by it.

What goes beyond this simple rhythm, in the rhythmical art of the Hebrews, amounts to but little. Here belongs,

1. The artificial arrangement of the alphabetical psalms. Thus Ps. xxv., xxxiv., xxxvii., cxi., cxii., cxix., cxlv., Prov. xxxi. 10 seq., the Lamentations of Jeremiah, with the exception of the last chapter, are alphabetically arranged by the initial letters of the verses, and this in different ways. Commonly each verse begins with a new letter; in Ps. xxxvii., however, only every other verse, though with interruption and change; in Ps. cxix. and Lam. iii., there are alphabetical strophes, as it were, that is, a series of verses have the same initial letters; in Ps. cxi., cxii., the half-verses are alphabetically arranged. This arrangement answers for us the valuable purposes of proving the existence of the parallelism of members, and of confirming the system

66

of accentuation in the division of verses and half-verses, respecting which we might otherwise have our doubts, as well as respecting the whole law of parallelism. The alphabetical arrangement is supposed by many *to have been intended to assist the memory. Michaelis, indeed, was of the opinion, that it was employed in the first place in the funeral dirge as an aid to the mourners, and afterwards employed on other occasions. Lowth supposes that the alphabetic poetry was confined altogether to those compositions which consisted of detached maxims, or sentiments without any express order or connection." I consider the alphabetic arrangement as a contrivance of the rhythmical art, an offspring of the later vitiated taste. When the spirit of poetry is flown, men cling to the lifeless body, the rhythmical form, and seek to supply its absence by this. In truth, nearly all the alphabetical compositions are remarkable for the want of connection (which I regard as the consequence, instead of the cause, of the alphabetical construction), for common thoughts, coldness and languor of feeling, and a low and occasionally mechanical phraseology. The thirty-seventh psalm, which is the most free in its alphabetical arrangement, is perhaps alone to be excepted from this censure, and in truth is one of the best didactic poems of the Hebrews. The Lamentations are, indeed, possessed of considerable merit in their way, but still betray an unpoetic period and degenerated taste.

In many of the alphabetic pieces we observe certain irregularities and deficiencies, which many (as Capell) have incorrectly imputed to the transcribers, who were the least exposed to commit mistakes in these compositions, since they were confined by the peculiar arrangement itself. In Ps. xxv. two verses begin with, none with ; yet the word in the second verse (like the interjection of the Greek tragedians uo) might not have been included in the verse, or (as Bengel conjectures) might have been written in the margin, in which case the following would restore the alphabetical order. Also in this, and in Ps. xxxiv., the is wanting; perhaps it should be restored by the

* As Lowth, p. 29, 259, and Michaelis on Lowth, p. 562, ed. Rosenm.

,מְצוּקוֹתַי

in the beginning of the second hemistich of the verse commencing with; and so also, perhaps, the p, which is wanting in the seventeenth verse of the former psalm, should be replaced by the P in at the beginning of the second hemistich. On the other hand, two verses begin with, and after the last letter, n, follows another D. This last we find also at the close of the thirty-fourth psalm. Michaelis supposes the is counted twice, on account of its double pronunciation, as Pe and Fe. Hasse erected upon it a paleographical hypothesis peculiar to himself, which is hardly capable of being sustained, and gives no satisfactory explanation of the phenomenon to be explained. According to this, the concluding, with a softer pronunciation, takes the place of the in the Greek alphabet. The conjecture of Bengel † is no better, who supposes that and both sprung out of the Phoenician Vau and Fau, and that the latter stands for the former; then the supernumerary verse with must come in the place of 1. Rosenmüller (1st edit.) considers both verses as the additions of a later hand, by which these psalms were prepared for the public service. But this could not be the case in respect to Ps. xxxiv. at least, as the last verse is necessary to the concluding of the whole; the conclusion of the twenty-fifth psalm is also very appropriate, and cannot well be dispensed with. In Ps. xxxvii., Y precedes D, y is wanting, and is repeated. Bengel accounts for this not unsatisfactorily from the interchangeable use of and y in Chaldee. Others resort for help to criticism. The thirty-ninth verse begins with ny, where perhaps the was not regarded. In Ps. cxlv. the verse with is wanting, which, according to Michaelis, has fallen out of the text. In Lam. ii., iii., iv., precedes y, which Bengel explains in the same manner as the similar fact in Ps. xxxvii. The order only is different; it was the custom to place letters of a similar sound together.

Another explanation of this irregularity is given by Vogel in Capelli Crit., T. I., p. 123.