Before Adjectives. 1. The classic poets extended the writing of the so-called masculine article to adjectives also beginning with accented a,19 but did not convert privilege into license. Although a number of cases can be cited from Lope's plays, it becomes relatively small when the great volume of his work is considered; and there is palliation in the fact that the peculiarity is virtually confined to the very common adjectives alta and ancha.20 La alta and la ancha having become familiar from much use, poetic reasons would explain and justify the slight variation. It is possible to find after prolonged search examples in modern poets, also.21

2. In writing el with alta and ancha the poet is within the bounds of the recognized usage of his time. But inasmuch as el is extended from nouns beginning with accented a to nouns beginning with unaccented a, we might expect a similar extension to adjectives under the same circumstances. Nor does the expectation prove vain, for Lope is equal to this extension, furnishing, however, only one example.22 The correctness of the passage is not open to question. The necessity for an additional syllable is probably sufficient explanation of the peculiarity.

II. WORDS THAT SUFFER APOCOPATION:

Algún mohatra o prestado 10. 15. 1. 34 is doubtless an inadvertence, as are also: E quebramos un costilla 5. 60. 1. 30 and:

19 Bello y Cuervo, l. c., 271 and 272 notes.

20 The remaining adjectives before which el might be written are: ágil, agria, alba, apta, alma, amplia, anua, ardua, áspera, aurea and hábil, Owing to their meaning and their position with reference to the substantive, the feminine article would seldom appear in immediate contact with any of these adjectives. The examples with ancha are: el ancha cava 6. 169. 1. 23; el ancha cuchilla 12. 133. 2. 11 (spoken by Pacheco) with áspera are: el áspera montaña 8. 138. 1. 16; el áspera grana 12. 162. 1. 26 and the remainder are with alta: el alta cerca 15. 475. 1. 11; del alta empresa 8. 219. 1. 12; al alta esfera 6. 457. 2. 1; el alta Alemania 2. 433. 1. 11.

21 Estalla al fin, y rinde el ancha copa. Andrés Bello, Silva, A la Agricultura de la Zona Tórrida.

22 El antigua amistad que tengo a un hombre 14. 142. 1. 11. Equally uncommon is the combination in the ballads, also. From an entire volume of the Romancero General may be quoted only: Por el alevosa muerte | Del rey Don Sacho su primo. Bibl. de Autores Esp., 10, p. 511, No. 790.

24

Poner un barda, un xamuga 5. 41. 1. 6. The apocopation 23 is intended to illustrate the ease with which Moors might wander from the straight and narrow path of gender in speaking Spanish. Hence these may be dismissed as studied cases of incorrect usage. Of a different character are several in which una is apocopated before alta. Yet the shortened form does not occur with ancha or áspera. A plausible explanation is that because of the very diffusion of alta there would be greater reason to introduce variations with it as a corrective of triteness. Examples, though nowhere numerous, can be reproduced in the ballads and even in more recent poetry written in the spirit of the ballads.25 It follows that un alta was current as a poetical phrase, correct because usage made it so.

It was a decided peculiarity if not an anomaly at any period to write este alma, whether in prose or verse. And yet this usage occurring in the earliest literary monuments maintained itself for a considerable time.26 Lope, however, displays some partiality for that writing. It is a question whether he deserves censure more than praise. The classic writers avoided as much as possible the use of a before another tonic a 27 and it has now become evident that Lope had an exceptionally sensitive ear for discordant combinations and in his effort to eliminate them was willing to prac

23 Of course examples abound of apocopation common at that time but not so usual now: algún ave 4. 4. 1. 9; la primer huerta 10. 99. 1. 45; la postrer sentencia 5. 118. 2. 38; un aya 10. 137. 1. 23.

24 A un alta cruz 13. 530. 2. 33; en un alta torre 6. 85. 1. 42; un alta sierra 13. 147. 2. 21.

25 Un alta roca, Romancero General 1, No. 405, p. 268; un alta empresa, No 521, p. 354; un alta torre, No. 836, p. 534; Duke of Rivas, Romances Históricos: Ve, al doblar un alta roca | Del faro amigo la estrella. (Colección de Autores Cas., 115, p. 53). I do not find this apocopation mentioned by any grammarian whom I have consulted. It should be a relic of the earlier period, yet it does not occur in the Poema del Cid where apocopation of una is rare even with nouns beginning with a vowel, e. g., almofalla and ora (Menéndez Pidal, l. c., I, par. 62, p. 233, and par. 49. 3).

28 Menéndez Pidal, l. c., I, par. 127; Salvá, Gramática de la Lengua Castellana, Paris, 1849, 350. 1.

tise extension beyond recognized limits.29 Other writers of the period, if the text be correct, have written este with feminine nouns beginning with atonic a, also.29 This Lope has not done.

IV. OTHER ADJECTIVES:

Of considerable interest and importance because of the tendency which they indicate are such expressions as todo el África, previously but only cursorily, mentioned. This coupling of masculine adjective with feminine noun owing to the juxtaposition of el appears sporadically in the literature only to meet with severest condemnation from purists.30 The examples 31 from Lope furnish as strong evidence as any could do of the careless workmanship of which at times he was guilty. They do more. They add their mite of proof to the fact that the word group toda el is unnatural and tends to yield to todo el. When this has happened, the noun has become for the moment masculine.

We finally come to the point which suggested the entire study. The only indication throughout Lope's many pages that an adjective immediately preceding and agreeing with a noun of the alma class could assume masculine form is contained in a passage 32 the uncouth barbarity of which renders serious consideration of its structure almost unnecessary. The expression itself is mocho agua 2. 191. 1. 4. The part is written to be spoken by a Morisco.

28 The list is only partial, of course: Ese alma 9. 602. 2. 45; ese agua 11. 118. 1. 21; este aspa 12. 309. 2. 41; ese hacha 5. 241. 2. 28; este águila 12. 602. 1. 8; aquel alma 11. 137. 2. 36; aquel Arca 4. 351. 1. 23. "Calderón, El Alcalde de Zalamea, 1, 654: Este ayuda; Menéndez Pidal, 1. c., par. 127.

20 Salví, l. c., 144. 7: sin que pueda sufrirse el alma atribulado ni

Mientras vuela risueño

El aura de la vida

como ha dicho Lista. Cf. Rivodo, l. c., Vol. 1, En. 11, VI, p. 95. See, also, Cuervo, l. c., par. 203 for examples drawn from writers of the New World.

Todo el alma se me abrasa 9. 426. 1. 40; Que dió a España majestad, | Y a todo el Africa miedo 12. 488. 1. 9; Valor tengo que puede a todo el Africa... Suceder en gobierno 14. 158. 1. 6. But todo is correct in: Todo el Alhambra baja 11. 157. 2. 7 and the example may point the way to one source of irregularity in the preceding phrases.

The lines are: Alea engarganta la olia | A vox que mete a esconderme, | Per no passar el tragonte | De mocho agua e mochos peces. Cf. Ticknor, Span. Lit., 2. 297, n.

Though no other example of disagreement occurs in the case of an adjective, petty errors of gender in the use of articles are found as well as errors in pronunciation and other evidences of imperfect acquaintance with Castilian Spanish. With no other example more significant than this before us, our conclusion on the point is reasonably safe. Lope wrote consistently mucha agua, poca agua, mucha hambre. If he did not go so far as to say mucho hambre, it is correct to infer that no expression 33 of this kind is to be found in any reputable author of the time.

University of Illinois.

H. M. MARTIN.

A FORGOTTEN LOVELACE MANUSCRIPT

Few seventeenth century lyrics are better known or more admired than Lovelace's To Lucasta: On Going to the Wars and To Althea: From Prison. On these two poems Lovelace's reputation rests. By snipping away undesirable stanzas, makers of anthologies have produced two or three other passable poems, but after all, no one cares much about The Rose, The Grasshopper, or To Amarantha: That She Would Dishevel Her Hair, even in their altered form. The extraordinary excellence of the poems To Lucasta and To Althea gives a special significance to any question involving the text of either one of them.

When Professor Schelling published his edition of seventeenth century lyrics in 1899, he printed lines 7 and 8 of To Althea as follows:

The gods that wanton in the air
Know no such liberty;

1

adding in the notes: "Gods. The original reading. There is no authority for birds, the usual reading." Others have followed him in printing gods instead of birds.2

* Mucho hambre is current among Chileans and perhaps the inhabitants of other South American countries. Cuervo, l. c., par. 203; Menéndez Pidal, l. c., par. 127; Roman, Diccionario de Chilenismos, Santiago de Chile, 1913, under hambre.

1 F. E. Schelling, A Book of Seventeenth Century Lyrics, p. 267.

2 W. C. Bronson, English Poems (The Elizabethan Age and the Purita

Schelling, to be sure, had, for the restoration of gods to the text, the authority of the first edition of Lucasta, which Wood says Lovelace himself prepared for the press. It is, however, to be questioned whether Schelling sufficiently considered the arguments that might be advanced for emendation. To begin with, birds pretty obviously makes the more sensible reading. Indeed it can hardly be doubted that Lovelace had birds rather than the Greek and Roman divinities in mind. Professor Grierson offers, in my opinion, the only possible defense of the text when he says: "The 'Gods' probably are the birds. Compare Aristophanes, The Birds. 11. 685-723, translated by Swinburne, Studies in Song." Yet it scarcely seems as though even so careless an artist as Lovelace would have wantonly destroyed the manifest parallelism between fishes and winds below by choosing to write gods instead of birds. And certainly the climax.

Angels alone, that soar above,

Enjoy such liberty

is seriously weakened by the previous mention of gods.

Hazlitt

Another argument for emendation was brought forward by Hazlitt as long ago as 1864 in his edition of Lovelace. called attention to a manuscript of the poem in the possession of Philip Bliss. This manuscript, which appears to be contemporary with Lovelace, was printed by Bliss in his edition of Wood's Athenae Oxonienses. Here the reading is birds instead of gods. Hazlitt felt that this manuscript evidence, coupled with many gross typographical errors in the original text, justified his printing birds. But Bliss himself had been somewhat doubtful, and others have apparently not been convinced.

There is, however, another hitherto unnoticed piece of evidence, which makes the case still clearer for the emended reading. In 1802 there appeared in the British Critic an anonymous review of Ellis's Specimens of Early English Poets. In a foot-note the

5

Period), p. 296; F. B. Snyder and R. G. Martin, A Book of English Literature, p. 116; H. J. C. Grierson, Metaphysical Lyrics & Poems of the Seventeenth Century, p. 61.

3 W. C. Hazlitt, Lucasta, p. 118.

4

London, 1813-20; 3. 461. The Bliss MS. is now in the British Museum (Add. MS. 22603, f. 16).

19. 621.