MA S TER NEGA TIVE NO. 92-80599-20 MICROFILMED 1992 COLUMBIA UNIVERSITY LIBRARIES/NEW YORK as part of the "Foundations of Western Civilization Preservation Project" Funded by the NATIONAL ENDOWMENT FOR THE HUMANITIES Reproductions may not be made without permission from Columbia University Library COPYRIGHT STATEMENT The copyright law of the United States - Title 17, United States Code - concerns the making of photocopies or other reproductions of copyrighted material... Columbia University Library reserves the right to refuse to accept a copy order if, in its judgement, fulfillment of the order would mvolve violation of the copyright law AUTHOR: NUTTING, HERBERT C TITLE: STUDIES IN THE SI-CLAUSE PLACE: BERKELEY DA TE : [1 905] COLUMBIA UNIVERSITY LIBRARIES PRESERVATION DEPARTMENT Master Negative # BIBLIOGRAPHIC MTrROFORM TARHFT ■■pi|l|« Restrictions on Use: Original Material as Filmed - Existing Bibliographic Record ID:iNYCG92-D3204a CC:9665 BLTram CP 5 cau PCss HMD I 010 040 100 1 2A5 :ia •**• %mif !«■/ "inn 490 L:eng PD:1965/ DCF XMT OR: RTYPsa CSC a GPC: PEP: POL s b 1 : p MOD : B 1 : CPIsO Dh RR: FRM2 F I C s (D FSI:0 COL n MS 05171c33 HNC= I cMNC Nuttinq, Herbert: Chester, =^ I dlB72- 1 934„ Studies in the si-clause^ 1 hCmicrof arm 1 .:--•• l. cBy H Berkeley , = I bUn i ver si ty Pre5<^~ j cf J 90"3 J p . C 35 1 -94 , = ( c27 cm . ' University of California publications EL. : u ATCs COM s b EML ■r AD 5 05 UD; 05 11:0 GEM: '07-92 ■07-92 BSE: C r. Muttinq. '.-•> • Clas B i cal p h i loloq y , v . 1 , no . 504 I n r 1. ti d e B ta i t) 1 i o q i-- a p f-, i c a 1 i'- e f e i*^ e r< c. e s I- ConcesBive si-clauses in Plautus dicative apodosis in PlautuB, 600 1 PI a u t u B , r i t: u s Ma c c i u s =:: I i,- L. a n n 1 4 a a i^ . o .3 O u L a t i n 1 a n q u a (:| e = I >; C J. a kx •?> * ;* •..:; « 650 O Latin lanquaqe- i xCcvi iunrtion'-. LD6 RL I M D 05-07--92 II. Subjunctive protasis with in TECHNICAL MICROFORM DATA FILM SIZE:___3l2r_Zn__ REDUCTION RATIO: / / ^ IMAGE PLACEMENT: lA (TfA) IB IIB DATE FILMED:__6__J_-_5l: INITIALS.,,"?^ ^ C HLMEDBY: RESEARCH PUBLICATIONS. INC WnODDRlDGE7cT «j: Association for information and image Management 1100 Wayne Avenue. Suite 1100 Silver Spring. Maryland 20910 301/587-8202 Centimeter 1 2 3 UlilllliiUMiUl^^ LU T 1 r Inches TTT 5 6 iliiiilniiliii 7 8 ilimliiiil I I I TTT imiiiiiiiiim I 10 11 nlimlmili 1 \\ 1.0 U^ 12.8 1.4 2.5 22 I.I 2.0 1.8 1.6 1.25 12 13 14 15 mm jplmjmjhmlmjlmi MfiNUFflCTURED TO fillM STPNDORDS BY RPPLIED IMRGE, INC. « UNIVERSITY OF CALIFORNIA PUBLICATIONS CLASSICAL PHILOLOGY Vol. 1, No. 2, pp. 35-94 January 31, 1905 STUDIES IN THE SI-CLAUSE I. CONCESSIVE SI-CLAUSES IN PLAUTUS. n. SUBJUNCTIVE PROTASIS WITH INDICATIVE APODOSIS IN PLAUTUS. BY H. C. NUTTING BERKELEY THE UNIVERSITY PRESS Price $o.60 UNIVERSITY OF CALIFORNIA PUBLICATIONS CLASSICAL PHILOLOGY.-Edward B. Clapp, William A. Merrill, Herbert C. Nutting, Editors. Price per volume $2.00. Volume I (in ' progress) : No. 1. Hiatus in Greek Melic Poetry, by Edward B. Clapp. Price, $0.50 No. 2. Studies in the Si-clause, by Herbert C. Nutting . . «' ' o.60 No. 3. The Whence and Whither of the Modern Science of Language, by Benj. Ide Wheeler (in press). The following series in Graeco-Roman Archaeology, Egyptian Archaeology, Ameri- can Archaeology aud Ethnology and Anthropological Memoirs are publications from the Department of Anthropology: GRAECO-KOMAN ARCHAEOLOGY. Vol. 1. The Tebtunis Papyri, Part I. Edited by Bernard P. Grenfell, Arthur S. Hunt, and J. Gilbart Smyly. Pages 690, Plates 9, 1903 Price, $16.00 Vol. 2. The Tebtunis Papyri, Part 2 (in preparation). BOTFnAH ARCHAEOLOGY. Vol. 1. The Hearst Medical Papyrus. Edited by G. A. Reisner and A. M. Lythgoe (in press). AMERICAH ARCHAEOLOGY AlTD ETHHOLOGY. Vol I. No. 1. Life and Culture of the Hupa, by Pliny Earlc Goddard. Pages 88, Plates 30, September, 1903 . . . Price, No. 2. Hupa Texts, by Pliny Earle Goddard. Pages 290, March, Price, Vol. 2. No. 1. The Exploration of the Potter Creek Cave, by William J. Sinclair. Pages 27, Plates 14, April, 1904 . . Price, No. 2. The Languages of the Coast of California South of San Francisco, by A. L. Kroeber. Pages 72, June, 1904. Price, No. 3. Types of Indian Culture in California, by A. L. Kroeber Pages 22, June, 1904 Price, No. 4. Basket Designs of the Indians of Northwestern California by A. L. Kroeber. Pages 60, Plates 7, January, 1905. Price, ' Vol. 3. The Morphology of the Hupa Language, by Pliny Earle Goddard (in press). AHTHROPOLOGICAL MEMOIRS. Vol. I. Explorations in Peru, by Max Uhle (in preparation). No. 1. The Ruins of Moche. No. 2. Huamachuco, Chincha, lea. No. 3. The Inca Buildings of the Valley of Pisco. 1.25 3.00 .40 .60 .25 .75 UNIVERSITY OF CALIFORNIA PUBLICATIONS CLASSICAL PHILOLOGY Vol. 1, No. 2, pp. 35-94 January 31,1 905 STUDIES IN THE SI-CLAUSE BY H. ('. MJTTING. I.— (H)\CESSIVE SI-CLAUSES IN PLAUTUS. In jrem-ral the hypotaetic concessive period may be defined as a complex seiiteiiee which brings together clauses of such a nature that tlie assertion in the conclusion might naturally seem to the hearer to he incompatible with the state of affairs referred to in the concessive chaise ; e.g.^ Rud. 1353 ff. ; Si maxume mihi ilium reddiderit vidulum, Non ego illic hodie debeo triobolum. Among the concessive periods of Plautus introduced by si and its compounds there is a large and striking class distinguished from the others by the (jrodc, so to speak, of the concessive clause. To diifei-entiate this group from what may be styled the simple (or normal) tyix^ I suggest the name "intensive." The simple type of concessive clause (as distinguished from the intensive) is characterized hy the fact that it goes no further than is de- manded })y the situation— it simi)ly recognizes a state of affairs (real or supposed) that has in some way been suggested to the mind of the s|)eaker: as, for instance, Men. 74() \X. ; Si me derides, at pol ilium non potes, I*atrem meum. Ps. 290 ff. : Egon patri subrupere possim (|uicquam, tam canto seni? Atque adeo, si facere possim, pietas prohibet. 36' University of California PuhlimUons. [Class. Phil. In neither of tbese passages does the eoneessive clause exceed the deaiaii^ nf the situation. In the irst ease the speaker refers to an ohvions fact when she says 8i me derides; for IMenaeehimis has heen treating her in a manner anything hut respectful. In thC' other lie .phw^ae m fmere posmm takes up a supposed possi- 'Hi mm- m .^ The intensive concessive clause on the other hand purposely exaggerate the state of affairs suggested to the speaker, e.g., Asin. 40B ff.. ; M. Atiine hercle ipanm adeo ci>ntnor: quassauti capite in- ecKiHc Quisque obviani hnic occesserit irato, vapulabit. MB. Siquideni hercle Aemeidink minis aniwmqm expktrn. eedit, Si raed: iratus tetigerit, iratus vapulabit. In tiii passage Hie mereatur might have heen content to confine himself to lie reported fact, thus producing a simple concessive period "Though he conies on in anger, he will get a beating if he touches^ jBi, ' ' But this ii^^'^lin taine m. laipfession for his emphatic mooi,. asi^'le liiai to tie extreme rftie improbable or iiiii)ossibIe —though Leoiiida comes nn (not merely angiy but) filled with lit boldness and courage ^f Aehillm, he will get a beating. An- utiif ' ^tm^ ■of lie same iind is CS. Qoaniiiaani tiM 8ii:8eensui, merh «na (U., «« w^) mrf ,uidem h«.„i„is; adfati.n. CA. Immo huic panimst. Nam si pro peeeatis reutiftH ducat uxoiis, pammst. Here irnfHm \evy obviously «'ai)s una of the preceding line. It is this gratuitous exaggeration that is the characteristic feature of the intensive type. C\>ncossive periods beloiitriu^^ to this cate- gory are generally easily i-eeognized when once the i)eculiarity of the type lias been noted, though of coui-se occasionally sen- tences are met with which are hard to classify. The intensive concessive peiiod is interesting from both tfie stylistic and the syntactical point of view. It is a form of speech common in dialogue, its most distinctive use being in emphatic Vol. 1] Nutting. — Studies in the Si-clause. 37 rejoinder. With Plautus it is a distinct mannerism ; about one- third of all the concessive clauses in his plays introduced by si and its compounds belong to this class. The lu-esent pa|)er deals particularly with tie syntactical asi)ects of the case. In the pages immediately following, the concessive periods introduced by si and each of its eompounds are pivsented separately, the sentences falling within the sevei-al groups being examined with reference to the distinction just made of simph' and intensive. Some points of minor interest are noted in passinix, but the more important questions suggested by the syntactical foi-m of the intensive concessive period are reserved for discussion at the end, after all the material has been presented. SI. It is (piite impossible to determine the exact number of con- cessive .'^/-clauses in Plautus. In many cases the nature of a clause depends upon the i)oint of view of the speakiM-, and there is no ()l)jective test by which to settle the (piestion definitely. Ex- cluding the more doubtful examples, I still find 88 .vZ-clauses that seem to diWrve the name concessive. This exceeds the estimate of Kriege,^ who puts the number at ()6. A.— Simple. Amph. 908; Si dixi, nilo magis es necpie ego esse arbitror. Capt. 12; Si non ubi sedeas locus est, est iibi ambules. (^ist. 27 ff. ; Si idem istud nos faciamus si idem imitemur, ita tamen vix vivimus Cum invidia summa. Mil. 681; Si albicapillus hie videtur, ne utiquam ab ingeniost senex. ]\Iost. 42 ff . ; Non omnes possunt olere unguenta exotica, Si tu oles. Rud. 1400; Non hercle istoc me intervortes, si aliam praedam perdidi. ' De eiiiintiatis concessivis apud Plautum et Terentium. Halle, 1884, p. 4. m Umversity of California Publications, [Class. Phil. Tri. 485 ft'. ; Semper tii hoe faeito, Lesbonice. eo^rites, Id optuiniim esse tute uti sis optiuniis. Si id neqiieas, saltern iit optimis sis proxiimiis. Tri. 507 ff. ; Sed si liai'c rt s jiraviter eecidit stiiltitia niea, Philto, est a«rer sub urhe hie nobis. True. 854 ft'. ; Blitea et luteast inervtrix nisi (piae sai)it in vino ad rem suam : Si alia membra vino madeant, eor sit saltem sobrium. For other eases see Asin. m:\ ft'., 9:^3, Aul. 254, Baeeh. 179, 365, 887, 1013 ft*., 1193 ff., Vi\\A. 223 ft'., 683 ft'.. 742 ft'., Cas. 298, 314 ft.. Cist, (w, 152 ft'., Ep. 599, Men. 670. 746, :Mere. 636, 819 ft\. Mil. 298, 3()<> tt, 747, Most. 914. Poen. 51, 374, I's. 290 ft*., Rud. 159, 1014, 1075, 1353 ft*., St. 43 ft'., Tri. 85 tf., 465, 607, True, 66 ff., 615, 833, 877. Total. 48. Some of these sim[)le eoneessive periods are a mere optional form of expression for a thouj^ht that miji^ht have been conveyed by two eoiirdinate clauses joined by an adversative conjunction. Such a case is IVIil. 631 (quoted above in full) ; there the speaker, had he so elected, mi^dit have expressed his thought in the fol- lowing form: **He looks gray, but in spirit he is by no means old.'' A more striking, and, at first sight, apparently unwarranted use of the form of a hypothetical concessive period appears in passages like True. 613 tf. ; STK. Verbum unum adde istoc: iam hercle ego te hie hae oft'atim oftigam. CV. Tange modo: iam ego te hie agnum faciam et medium distruneabo. Si tu ad legionem bellator clues, at ego in culina elueo. Jn this last line the form of the first clause is easily justified, but the words at (t/o in culina clufo, taken at their face value, ♦lo not complete the meaning of a concessive period. There is, it is true, an antithesis between the two clauses ; but a genuine con- cessive period involves something more than mere antithesis — Vol. 1] Nutting. — Studies in the Si-clause. 39 there is an incompatibility between the subject matter of the two clauses such that the hearer is surprised at the statement in the conclusion ; for the state of aft'airs here mentioned would natu- rally seem to be precluded by that referred to in the concessive clause : as in the typical sentence first quoted, Rud. 1353 f¥. ; Si maxume mihi ilium reddiderit vidulura, Xon ego illic hodie debeo triobolum In the sentence under discussion, as it stands, this element of incompatibility appears to be lacking; whatever the amount of warlike fame possessed by Stratophanes, there is nothing what- ever surprising in the claim of Cuamus that he is a famous per- former in the kitchen. If we must take the words at ego in culina clueo at their bare face value, the probable explanation of a sentence of this sort is that the line between simple antithesis and antithesis with incompatibility is not always sharply drawn ; in this way it might occasionally happen that clauses w^hich were merely antithetical would be strung along in the form of a hypotactic concessive sen- tence. On the othei- hand, it is quite possible that in the con- clusion of a sentence like True. 615 the speaker is not expressing himself fully, and that the underlying thought contains all the elements of a genuine concessive period. For instance, the mean- ing in this particular case might be "Though you are famed for valor in the army, (you need not try to frighten me, for) I am a famous performer in the kitchen." In the line that precedes the passage fjuoted, Cumanuis has shown that his performances in the kitchen include the handling of knives, thus helping us to fill out what (if this interpretation be correct) he leaves unex- pressed in 615. This second explanation is a very attractive one, and is the more justified because such abbreviation in verbal expression as is here assumed is no rarity in language generally.^ With True. 615 may be compared Bacch. 364 fl*. ; Si ero reprehensus, macto ego ilium infortunio: Si illi sunt virgae ruri, at mihi tergum domist. = American Journal of Philology, XXIV, p. 294. Cf. Lindskog, De enu- tiatus apiul Plaiitum et Terentium condicionalibus, Lunrlae, 1895, p. 103 ff. 40 University of California Publications. [Class. Phil. Baech. 885 ft*. : Quid illuin iriorte territas? Et ejro te et ille niactaiiiiis infortiinio. Si tihist inaehat'ra, at n(>!)is veniinast donii. Rud. 1014; Si til proreta rsti navi's, oiro Lriiboriiator (To. B.— Intensive. Tlu* iriost strikinir thiiiir {d)<)ut the examples that fall under this ln'adinjr is that, in more than half of the eases, the intensive force eentei's around some other word (or phrase) than the verh. As in the fon<>\vin«r; Asiii. 413 ff. : LI. liic me moratust. liH Si«juidem herele nune summutn hnuiu te dieas detiruiisse Atear also in the conclusion; e.g., St. 287; Si rex obstabit obviam, regem ipsum prius pervortito. ETSI. 26 cases. A.— Simple. In the examples that fall under this heading the nature of the sentence is generally so evident that it will be sufficient to quote only the cfi/clauses, omitting the conclusions. Aul. 421 ; etsi taceas. Bacch. 1160 ; etsi . . . prope scire puto me. Bacch. 1191; etsist dedecori. Capt. 543 ft'.; etsi ego domi liber fui,Tu. . . servitutem servisti. Capt. 744; etsi aliter ut dicam meres. Capt. 842 ; etsi nil scio quod gaudeam. Cas. 95S; etsi malum merui. Mil. 407; etsi vidi. Mil. 532; etsi east. Most. 6(}C); etsi procul abest.® :\lost. 854 ; etsi non metuendast. Pers. 272; etsi properas. IVrs. 601 ft'.; etsi mihi Dixit . . . Pers. 655 ; etsi res sunt f ractae. Poen. 1084 ; etsi hie habitabit. Ps. 1113; etsi abest. Rud. 1044; etsi ignotust. Rud. 1350 ; etsi tu fidem servaveris. Tri. 383 ; etsi advorsatus tibi f ui. Tri. 474 ; etsi votet. Tri. 527 ; etsi scelestus est. Tri. 593 ft'. ; etsi admodum In ambiguost . . . Tri. 600; etsi odi banc domum. True. 815 ; etsi tu taceas. B.— Intensive. There remain but two cases to come under this head; both belong to the second type of intensives described, i.e., the^verb is the center of intensity or else the intensity is distributed through- out the clause. * In the edition of Goetz and Schoell this line is placed between 609 and 610. vx University of California Publications. [Class. Phil. Vol. 1] Xutting. — Studies in the Si-clause. 45 Capt. 854 ff. ; HE. Xec nil hodit* nee nnilto plus tu hie edes, ne fnistra sis: Proin tu tui eottidiani vieti ventreni ad nie adferas. ERG. Quin ita faeiain, ut tute eupias faeere suniptum', etsi etro veteni. Vid. lOfi ff.; malo hunc adlijrari ad horiani Ut semf)er piseetur, etsi sit tenipestas maxima. In passin«r, the exeeedinj; hrevity of the r/.v/-ehiuse may be noted; 20 of the 26 elauses do not exeeed four woi-ds eaeh. TAMETSI. 16 eases. A. — Simple. For the eases that fall under this headinir the material mav be presented in the same way as foi- etsi, Aniph. 21 ft*. ; tametsi . . . Sei!)at. Amph. 977; tametsi praesens non ades. Aid. 768; tam etsi'* fur mihi's. Capt. 321 ; tametsi nniens sum. Cure. 259; tam etsi non novi. Cure. 504; tam etsi nil feeit. Mil. 744; tam etsi dominus non invitus patitur. Pers. *^62; tam etsi id futurum non est. Poen. 342; tam etsi in abstruso sitast. Poen. 1201 ; tametsi sumus servae. Ps. 244; tametsi oeeupatu's. Ps. 471 ; tam etsi tibi suseenseo. St. 41; tam etsi's luiiior. St. 205 ; tam etsi herele . . . iudieo. B.~Intensive. Men. 92; Numqiiam herele effiisriet, tam etsi capital feeerit. Tri. 679 ; Facilest inventu: datur ijrnis, tam etsi ab inimico petas. T]iou«:h the number of intensive cases is the same as for etsi. the smaller sum total rendei-s the proportion larg:er. There is also the further difference that these cases are of the type first described — the intensive force centers about some other word or words than the verb. Thou^^h not so strikinjr, the brevity of the tam€tsi-i^\i\UHe also desei'ves notice; 10 of the 16 examples do not exceed four words. ETIAM SI. 2 cases."* Ep. 518 ff*. ; immo etiam si altcrum Tautum perdundumst, perdam potius quam sinam Me inpune irrisum esse. Ps. 626 ff. ; PS. ^lihi herele vero, qui res rationesque eri Rallionis cui-o, arjrentum accepto et quoi debet dato. HA. Si (|uidem herele etiam supremi promptas thensauros I avis Tibi libellam arjrenti numquam credam. Both these cases are intensives of the first type — ^the intensive force centers elsewhere than around the verb. In the second case the resolution si . . . etiam is precisely papallel to ct . . . icat and " If . . . even ;'Mn translating the sentence the last named phrase mijrht be used to advanta«;e. In general, intensives of the fii-st type (however introduced in Latin) can be rendered by ** Though . . . even'' and '*If . . . even;'' in this way we have something more than stress of voice to mark the center of intensive force. In view of the very restricted and clearly defined use of etiam si. it is inexact, when dealing with the language of Plautus, to make the phrase si = etiam si a substitute for saying that a given ^/-clause is concessive. Sonnenschein makes such a note on Rud. 1400 ; Xon herele istoc me intervortes, si aliam praedam perdidi. The real parallel to this ^/-clause is the ^f5«-clause, as will be at » Tttmetsi is here written as one word or two, ac(M>r.line to tlie reaaintr of the Goetz-Schoell edition. ** '" Cas. 806 also shows the combination etmm si, but the passage is mani- festly cornii)t. 46 University of California Publications. [Class. Phii* cnce evident if it be compared with the exainples r|uoted uiuler that headinjr; the parallelism is coiiipkte, even to the nmnhor of words in the clause. TAMEN SI." 2 cases. Cas. 795; Qui amat, tanien herele si esnrit, nulhiiii r^surit. St. 27 ff . ; Tanien si faciet, niinunie irasei Decet : necpie id immerito eveniet. Both of these are simple concessive i>tM'i7; TH. Verberihns, lutnni, caedere pendens. TK. Tanien etsi pudet ? These two cases are also simple conci'ssive periods. It now remains to consider two jreneral syntactical peculiari- ties broup:ht to lijrht by a division of concessive clauses accordiniz as they are simple or intensive. First as to introductory particle ; the usa«.^e of Plautus can be conveniently examined in the follow- inir summary. si etsi tametsi etiam si tanien si tamen etsi 2 Totals Simple 48 24 14 2 90 Intensive 40 2 2 2 46 Totals 88 26 16 2 2 2 136 It will be seen that ,s/ has been used as the introductory particle in 48 of the 90 simple concessive periods, and in 40 of the 4(3 " Kriege (1. e.) docs not recognize this compound. "Kriege (1. c.) wakes this number four b\ including Cas. 9.5S and Poen. 1084. These cases are here enumerated under etsi, tamen being assigned to the apodosis. Vol. 1] Nutting. — Studies in the Si-clause. 47 intensive. The overwhelming preponderance of si in sentences of the intensive type presents an interesting problem. Appar- ently the key to the situation lies in the fact that, from the subjec- tive point of view, concessive clauses in general fall into two dis- nnct categoi'ies; by the use of such a clause the speaker may (a) concede that a thing is really true, or (b) concede it for the sake of argument, or the like. As in the following examples: (a) Cas. 957 If.; vapulo herele ego invitus tamen, Etsi malum merui. (h) Hacch. 1004; Nam ego non latui'us sum, si iubeas maxume. Bacch. 128; Qui si decem habeas linguas, mutum esse addecet. In the first of these passages etsi malum merui is scarcely more than a statement of fact; without altering the sense it could be made such by so rearranging the sentence as to give it first place. Hut in the cases that fall under (b) there is a totally different state of affairs; the concessive clause is a mere supposition, and, as such, is closely allied to the pure conditional clause ;^^ for in both the speaker is equally lacking in assurance of realization in fact. Therefore if si — a word whose distinctive function it is to introduce pure conditional clauses — is also to do duty anywhere as a concessive particle, clearly it is in concessive clauses of this second variety that we should expect to find it most freely used — and such in fact is the case. The overwhelming preponderance of si in sentences of the intensive type is but an illustration of the workings of this general principle; for in them the conces- sive clause by its very nature is a mere supposition — its essential characteristic being that it far exceeds the facts of the case, often flying to the extreme of the improbable or the impossible ; e.g., Asin. 414; Siquidem herele nunc summum lovem te dicas detinuisse. Aul. 100 ; Si Bona Fortuna veniat. Bacch. 697 ; Quern si orem ut mihi nil credat. ]\rcn. 751 ; si avom vis adducere. in Cf. American Journal of Philologj', XXIV, p. 279 ff. 48 University of California Publications, [('lass. Phil. Looked at from this point of view, the larj^e use of si in sentences of the intensive type ceases to be snrprisinjr. The fact that si introduces 48 of the 90 simple concessive periods does not perhaps seem to Crill so loudly for explanation, but it may be noted in passinjr that this ratio completes the illus- tration of the preneral principle above noted with reference to the use of si. A simple concessive period may be of either of the varieties above desi^rnated as (a) and (b). Si introduces prac- tically all that are mere suppositions, and has found its way to a considerable extent into clauses that admit a fact, leavinir the Iar*?er share of these latter however for its more distinctively conci*ssive compounds. In the followintr table the conccssivi clauses are a«rain classi- fied, this time with reference to the mood of the verl). The totals differ sli«rhtly froni those of the other table because, for the pres- ent purpose, it was necessary to exclude doubtful forms, such, for instance, as those in -am. 8i 30 15 etsi tametsi I etiam si i taiiien si tamen etsi , Totals Simple ( indie. Isubj. 20 2 14 2 •> 08 17—85 Intensive ^ r indie. l sul>j. 10 29 1 2 12 32—44 Totals 84 24 15 2 2 2 129 It here appears that in sentences of the simi)le concessive type the proportion of indicative to subjunctive is 68:17, while for the intensive type it is 12:32. The reason for this remarkable variation is doubtless to be found alontr the line of the distinction just drawn between those concessive clauses that admit a fact and those which are mere suppositions. A concessive clause that admits a fact is closely akin to a statement, and naturally takes the indicative;** whereas those which betray a lack of assurance about realization in fact (and are thus closely allied to conditional speaking) use soiuetimes one m(X>d, sometimes the other— nuich as so many conditional clauses mijrht do. From this point of view the meaning of the ratio of indicative to subjunctive Vol. 1] Xiitting. — Studies in the Si-clause. 49 (68: 17) for the simple concessive periods begin to appear; for, as above noted, we have here to do both with clauses that admit a fact and with those that are mere suppositions. The (very numerous) cases that admit a fact count solidly on the indicative side of the ratio, whereas the mere suppositions contribute a rea- sonable number to each member of the proportion. Under these circumstances a heavy preponderance of the indicative is just the thing to be expected in the totals.** The ratio of indicative to subjunctive (12:32) in the intens- ive periods cannot be explained so simply. Of course, we should expect to find both moods fairly well represented, for (as already shown) the intensive concessive clause is by its very nature a mere supposition, and would therefore in general follow the rules for mood in pu»^e conditions. But this is not a full explanation of the ratio 12 : 32 ; for in conditional sentences Plautus uses the indicative on the average much more frequently than he does the subjunctive. The intensive concessive clause however is some- thing more than a mere colorless supposition — it is generally a very wild and improbable one. Apparently it is this peculiarity that turns the scale so heavily in favor of the subjunctive.*" '* I am speaking here only of the language of Plautus, and in particular of the concessive clauses intro-luced by si and its compounds. Such a state- ment would not of course apply to a developed construction like the sub- junctive c«m-clause in concessive periods of Cicero's time. " In this connection it may be noted that the ctsi- and tametsi-clsiuaes ahnost uhvays concede a fact. The conventional rule for mood with these particles quite disregards this basis of explanation for the use of the indic- ative. ^^ This point is further considered in the following paper. See p. 88 ff. 50 University of California Puhlications. [Class. Phil STUDIES IN THE SI-CLAUSE. II.— STTBJrXCTIVE PROTASIS WITH IXDICATIVE APODOSIS TX PLArTlS.i In this paper the phra.se **snbjnnetive protasis with indica- tive apodosis*' is used in the l)r()ad sense in which it is connnonly iindei*sttK)d — that is, as includin*r all sentences whose subordinate clause chances to be introduced by si, irrespective of the exact nature of the underlyinir thou*rht. The ar»;unient throu«rhout is based on sentences which employ undoubted forms of the sub- junctive and indicative; those containing forms in -fiw, -nr, etc., could only brinjx an element of uncertainty into the discussion, and the material fortunately is abundant without them.- l.-PURE CONDITIONAL SENTENCES. As a preliminary to the detailed study of the sentences of this i^roup, attention may i)roperly be called to the somewhat undeveloped state of the language in the time of Plautus. With regard to this two points are of interest for the present discus- sion. In the Hi"st place, the uses of the subjunctive and the indica- tive were not in general so carefully differentiated as at a later period. For example. Ha me di atnahuuf and ita mc amabii lup- piter are used freely alongside of //(/ mv di amvut. Again, take the deliberative question : * See the Classical K'view, Vol. XVII, p. 449 If., for a critique of the work of Lilie, Lebreton ami Lodge aiKl the later theory of Blase on this subject. Blase 's earlier view will be found in De modorum temi)orumque in enuntiatis eondicionalibus Latinis permutatione, Dissertationes Philo- logicae Argentoratenses, Vol. X, p. 94 (38) ft'. Cf. Langen, Beitriige zur Kritik unecause the subjunctive of the .vi-clause niav be due to dependence on an infinitive or the like; Amph. 675, Aul. *Jl»8, 320, Bacch. 1193 ff., Ps. 1033 flF., and St. 112. Vol. 1] Nutting. — Studies in the Si-clause. 51 Ter. Phor. 736-37 : CH. Quid ago? 80. (pii est eius pater. CH. Adeo, manco, dum ... co- gnosce ? Ter. And. 639 : Sed quid agam? adeamne ad eum et . . . expostulemP Finally might be eited cases of remarkable variation of mood in conditional sentences; e.g.; Ps. 1070 ft'. : Roga me viginti minas, si ille hodie ilia sit petit us muliere sive eam tuo gnato hodie, ut promisit, dahit. Amph. 703 ff. : Bacchae bacchanti si velis advorsarier, ex insana insaniorem facicSy feriet saepius : si obsequare, una resolvas plaga.* It is possible that the very considerable middle ground afforded by forms belonging to both the subjunctive and the indicative (e.g., those in -am, -ar, -eris, etc.) tended to delay a sharp differ- entiation between the uses of the two mood systems. In the second place, in Plautus' day grammatical conceptions were neither so symmetrical nor so clearly defined as at a later time. This is shown in an interesting way in such contrary to fact sentences as the following : Aul. 523-24: Compcllarem ego ilium, ni metuam ne desinat memorare mores mulierum : nunc sic sinam. Bacch. 635: PI. Si mihi sit, polliccar. MX. Scio, dares : novi. Poen. 1251-52: primum, si id fieri possit, ne indigna indignis dei darent, id ego evenire vellem. St. 510-11 : Voce^n ego te ad me ad cenam, f rater tuos nisi dixisset mihi ' So also quid ago? and quid agam? in Ter. Phor. 447 and Hec. 715. * With these might be compared Cist. 683 ff. (si nemo praeteriit, iaceret) and Rud. 744 (iam tanta esset, si vivit). 52 University of California Publications. [Class. Phil. te apud se cenaturuni esse hodie, i{\unn me ad se ad eeuam vocat. True. 830 : Nam vinum si fabulari jjossit, se deft udc ret, ^ SiK-h combinations of forms as here oeeur we can readilv under- stand, for just at this tiine the contrary to fact idea was discard- in*,' the present (and perfect) subjunctive, findint; in the imper- fect and pluperfect a more distinctivi' and satisfactory form of expressi(m; but that the two forms should be min«rled within the limits of a sinji:le sentence betrays a lack of keen appreciation for synnnetrical sentence structure. These two characteristics of early Latin distinctly favored the freipient occurrtMice of subjunctive protasis with indicative apodosis. For the failure to differentiate clearly between the use of subjunctive and indicative forms in «;eneral must have affected also the choice of mood in the clauses of conditional sen- tences — in some cases, so far as meanin*; is concerned, there was doubtless little to choose between the two moods: and to a writer whose ideiis of symmetrical sentence structure were somewhat undeveloped the pairini; of different moods in the clauses of a conditional sentence probably did not appear to be such a strik- ing? irret^'ularity as it seems when viewed from a later standpoint." The many examples however in which Plautus uses the same m fi°d which led Langen (1. c. p. 48) to reject the line. ^ctu^cu Vol. 1] Nutting. — Studies in the Si-clause. 59 Capt. 259-60 : Xeque pol tibi nos, quia nos servas, aequomst vitio vortere neque te nobis, si abeamus hinc, si fuat occasio. Capt. 850: Sets bene esse, si sit unde. Capt. 906 : Nam si alia memorem quae ad ventris victum conducunt, morast. Cure. 299: Recte hie monstrat, si imperare possit. Men. 760 : quas si autumem omnes, nimis longus sermos^. Merc. 497 : Mcliust, sanus si sis. :\rei-e. 692-93 : Paiumne est malai rei quod amat Demipho, ni sumptuosus insuper etiam siet? :\ril. 1263 : Non edepol tu ilium magis amas quam ego, mea,si per teliceat. l\)en. 921 : nunc si eadem hie iterum iterem, inscitiast. Ps. 740 : Quid? si opus sit ut dulce promat indidem, ecquid habetf St. 171-72: Nunc si ridiculum honiinem quaerat quispiam, rtnalis ego sum cum ornamentis omnibus. Tri. 557-58 : Quin hie quidem cupit ilium ab se abalienarier, siquem reperire possit, quoi os sublinat. Here again, especially when the apodosis precedes, it is difficult to say with certainty under which of the three heads a given example should be classified. However, Ps. 740 and St. 171-72 seem clearly of the first type, i.e., the state of affairs referred to in the apodosis is felt as in no way dependent on the truth of the condition. (b). Sentences of the form si sit— erit. Asin. 699 : Vches pol hodie me, si quidem hoc argentum ferre speres. 60 University of California Puhlicaihnis. [rLAss. Phil Vol. 1] y Kiting. — Studies in the Si-clause. 61 m Aul. 311 : Faniern herele iitendaiii, si rojres, nuuKinani rlabif. Cure. 18fi: Irmcere, si te edeiiteni hie a cibo abijrat. Merc. 650-51 : Si i})i amare forte oeeipias atfjue item eius sit inopia, iaiii inde porro aufugiis/ Mil. 571 : Ne til herele, si te di anient, linjriiam compritnrs. Most. 56-57 : Ita te forahunt patihulatuin per vias fstiinulis, si hue reveniat senex. F*oen. 729 : Si pultem, non reel tide t, Poen. 1085: Qiiin niea (iiuxjue iste habebit, si(|iiid me fuat. Tri. 26-27 : Conrasfif/abo pro eommerita noxia, inritus, ni id me invitet lit faeiam fides.^- Por the purposes of the present diseii-jsion the last example cited may he i«rnoi*ed because the apotlosis is really invitus rathei* than Coneastifjabo. The most strikinjr thinjr about the ^'roup is the prevalence of sentences of the seconti type, i.e., sentences in which the apodosis refers to a state of affairs felt as conditioned. Cure. 186, Merc. 650-51 and Poen. 729 fas here punctuated^') are clear cases. So apparently Asin. 699, Most. 56-57 and Poen. 1085, unless the first be a case of anacoluthon. In the sentences of other forms thus far dealt with the explanations for examples of the second type have been the modality of the verb of the apodosis and the undeveloped state of the lanjruajre. Here how- ever none of the verbs are modal, and we are ajrain forced back (as in the case of Cas. 528-29) to the other line of explanation. But in this cateofory the easy tolerance of the unsymmetrical sen- tence structure is much more readily understood. For the verb of the apodosis refers to the future — a time realm in which the bounds of indicative and subjunctive mcaninor were perhaps least "The manuscript reading would add Baeeh. 1172 to this list. "The more difficult punctuation is si pultem, uon rechidet? i.e., ''What if I knock and he does not open?'* clearly set in early Latin. Plautus perhaps felt it no harsher to use the futures of ordinary verbs in this way than to so employ the presents of modal verbs. If so, we can readily understand the prevalence of sentences of the second type in this catejiory. Of the two cases of this form not yet treated, Aul. 311 seems of the first type, the action of the apodosis being independent of the truth of the protasis. The other case (Mil. 571) has no parallel am(>n*r the sentences thus far treated, the future indica- tive havinjr somethin»r of imperative force. The whole passage is PE. Ne tu herele, si te di anient, linguam eomprimes : posthac etiam ilhid (piod sceis neseiveris nee videris quod videris. SC. Bene me mones. The line here between indicative and subjunctive was not very clearly defined, as we may see by comparing line 293 of the same play : Verum etiam tu istam, si te di anient, temere hau tollas fabulam. (c.) Sentence of the form si fuerim erit. Cas. 335 ff. : Sed tandem si tu luppiter sis emortuos, (piom ad deos minoris redierit regnum tuom, quis mihi subveniet tergo aut capiti aut cruribus? This also is an example of the second type. (d.) Sentences of the form si esset (fuisset)- Amph. 947-48: T't quae apud legionem vota voviy si domum redissem salvos, ea ego exsolvam omnia. Bacch. 818-19; Ilunc si ullus deus amaret, plus annis decern, phis iam viginti mortuom esse oportuit. Cas. 440-41 : Volui Chalinum, si domi esset, mittere tecum obsonatum. Mil. 475-76 : Quid propius fuit quam ut perirem, si elocutus essem ero? -fuit. .^.^.^^athm^Mt^m MiiMifnMii. iA.^^ iM lUilMMJiwiM^Hiku 62 University of California Publications. [Class. Phil, Mil. 1356-57 : et si ita sen tent ia esset, tibi servire mahd multo s be inclined to count this another example of the second " I'f. Tri. 8.H5 ff., which may be so punctuated as to form a parallel. " C'f . Ps. 499. uiiaiailiiiiiiduitLuiiiiyiii'r uj iLkt Vol. 1] Nutting. — Studies in the Si-clause. 63 type — decidi would then be a mere piece of exaggeration. But paene disqualifies hs clause for being the apodosis of ni hie ades- ses: for the realization in fact of that condition would have meant actual falling in. not almost falling in. The phrase ego ille doc- tus leno paene in fovcam decidi is therefore worded without ref- erence to the addition of the ^/-clause at the end, and as a matter of fact it is in itself a complete and precise statement needing no further qualification. In other words we seem to have to do with a case of anacoluthon, but this is different from any exarii- ples of the phenomenon yet taken up. A comparison of the fol- lowing sentences will make this point clear. Xe] ego a ma re utramvis possum — si probe adpotus siem. Omnia istaec scimus iam nos — si hi spectatores sciant. Ego . . . paene in foveam decidi — ni tu adesses. In the first of these examples the speaker corrects the opening lemark by the use of the sZ-clause, letting the hearer know that the state of aiiairs there asserted is after all subject to a condi- tion. In the second the si-clause is added as a necessary check on the hearer's unconditioned inference from the statement Otn- nia istaec scimus iam nos, namely **you need not enumerate them. ' ' In the last example neither of these things is true ; paene in foveam decidi and the obvious inference to be drawn from it ("I did not fall in") are both facts subject to no condition, and neither therefore needs a corrective w«-clause; and such is not the function of ni hie adesses. Rather, this contrary to fact phrase is used to imply the reason why the speaker did not fall into the trap. Without makinir any elaborate analysis it is clear that this implication is the chief function of the clause; for the speaker is obviously using the words to express his obligation to the hearer for his presence (and advice), representing them as the cause of his escape. In other words, ni hie adesses does not cor- rect the preceding statement or the unconditioned inference from it C'l did not fall in"), but it further extends the thought of the sentence by assigning the cause for the thing to be inferred. It is customary to treat sentences of this sort as the result of ellipsis, but the above analysis suggests another possible line of explaiuition. In Plautus there are many regularly formed con- PI 64 University of California Publications. [Class. Phil. Vol. 1] Xntting. — Studies in the Si-clause. 65 trary to fact conditional sentences whose chief function is to assij?n a reason for an existing or past state of affairs; e.g.. Mil. 12H2: MI. Non video. I'hist ? AC. Videres pol, si aniares.^® In this passajre Videres takes co«rnizance of the fact stated in the precedinjr speech (Sim video), and the .s/-clause assi«rns the rea- son for that fact, i.e., that the fii*st speaker is not really in love. In the sentence under discussion ( paeue in fovtam deeidi, ni Jfie adesses), at the end of the fir*st clause the speaker may become conscious that his words take cotrnizance of the fact that he did not fall in, just as would have been the case had he said (heidis- xf //^ and this p«'rhaps tempted him to use, in acknowlcdirin«r the cause of his not fallin*; in, the form which is »renerally employed only whiMi the fact for which a reason is assitrned is implied by a contrary to fact subjunctive apodosis, as in Mil. 12()2 above. (e). Sentences of the form si esset fuerat (erat.) Baech. 563 ff.: Quid? tibi non erat nieretricum aliannn Athenis copia, (juibuscum halxMvs reuK nisi cum ilia (juam cuo mandass(»m tibi, (K'ciperes tute etiani amare . . . f Mil. 52-53: Quid in Cappadocia, ubi tu (|uin*rentos sinnil, ni hebes nuichaera foret, uno ietu o'cideras? St. 512-13: Et niajris i)ar fur rat me vobis dare cenam advenientibus, qiuini nie ad ilium promittere, nisi nollem ei advoi*sarier. The interesting: example of this jrrouj) is Mil. 52-53, showin»r as it does the same sort of /< /-clause as appears in Peis. 594-95, which has just been discussed at lenjrth. The explanation here however is much easier, for the ///-clause precedes,'" and the action re- ferred to in the apodosis obviously depends on the cominj; to pass of the condition that was not realized. This therefore is but another example of the second type, and is to be explained partly " The other cases are enumerated in the Anierit^an Journal of Philology, Vol. XXII, p. 310 ff. " This precludes treating the sentence as a case of anacoluthou. in the same way ;).■; others of that class, partly on the jrround of the spirit of exairireration that pervades the passag:e in which the sentence occurs.'® The other two cases in this «:roup seem also to be examples of the second type; St. 512-13 has a modal verb in apodosis. In snmniintr np the results of this study with reference to j)ui'e conditional sentences, it may be remembered that we began with the assumption that Plautus had a fairly clear conception of the same mood in both clauses as the norm. The problem in hand is therefore to discover the I'casons why some sentences do not conform to that norm. P^our such reasons have been enu- merated. 1. The fact that the state of affairs mentioned in the apodosis is often in no way dependent on the truth of the protasis; the indicative statement includes and implies what would be in the supposed case. 2. The modal meaninj; of certain verbs, notably posse. 3. The union of a complete sentence and a part of another by anacoluthou. The form of each member of the expression is de- termined by the thought it is to convey, irrespective of the form of the other member. 4. The somewhat undeveloped state of the language in Plau- tus' day, as shown (a) in irregular sentence structure and (b) in the not very precise use of mood forms. This method of expla- nation finds its most sweeping application in cases referring to the future: for there the fact that the realms of indicative and subjunctive meanings were not carefully differentiated tended to make the lack of symmetry in sentence structure still less noticeable to Plautus than it would otherwise have been. Aside from sentences referring to the future there are very few IH Wit With regard to such sentences as this it should be remembered also that at this time Latin was in the midst of the process of adopting the use of the secondary tenses for the expression of the contrary to fact idea. In Greek it was the indicative that was chosen when a similar shift of tense was made in that language, and it is possible that we should recognize in early Latin some sporadic and unorganized impulses to develop in that way rather than toward the use of the subjunctive. Cf. Men. 195 (si amabas), regard to the case under discussion Brix seems to lay too much stress on the demands of the metre; cf. his note ibid. 131. Hi' m University of California Publicatiom. [Class. Phil. cases for which this is the only possible line of explanation. Gen- erally it is to be combined with others, as for instance with 2 above. One or two combinations with factors not here enumer- ated were mentioned in the discussion of individual cases. ^* IL-CONCESSIVE SENTENCES. (a). Sentences of the form si sit est. Asin. 318-19: Si quidem omnes coniurati cruciamenta conferant, habeo opinor familiarem tergum, ne quaeram foris. Asin. 933 : Pol si aliud nil sit, tui me, uxor, pudet. Bacch. 128: Qui si decern habeas lin^aias, mutum esse addecet. Bacch. 1045-46: Si plus perdundum sit, periisse suariust quam illud flajjitium vol«:o dispalesccre. Cas. 314 ff. : Quin si nolis filiusque etiam tuos, vobis invitis atque amborum ingratiis una libella liber possum fieri. Cist. 27 ff . : Si idem istud nos faciamus, si idem imitemur, ita tamen vix vivim us cum invidia summa. Merc. 841 : Ibi quidem si re*rnum detur, ncm cupitast civitas. Pers. 40-41 : Quin si ejromet totus veneam, vix recipi pofis est (|uod tu me rogas. Ps. 291 : Atque adeo, si facere possim, pietas prohihet. St. 43 ff . : Et si illi improbi sint . . . nostrum officium meminisse decet. Vol. 1] Nutting. — Studies in the 8i-clause. m "The jussive force of the future indicative (Mil. r>71) and the exag- geration which pervades the passage in which Mil. 53 occurs. Tri. 1186 : Nam si pro peccatis centum ducat uxoris, parumst. True. 877 : Factum cupio : nam nefacere si velim, non est locus. The sentences of this group well illustrate the tendency of the concessive ^/-clause to precede its conclusion ; here there is not a single variation from the rule. Excepting in Cist. 27 ff. and possibly in Merc. 841 the ^/-clause is a mere supposition, and takes the subjunctive mood for the same reason that that mood is employed in pure conditional sentences of the ideal and con- trary to fact types. Having begun his sentences with such a si- clause. Plautus nevertheless does not hesitate to complete them with an indicative conclusion, and such a course is not without justification. For in the above examples it will be found that the conclusion refers regularly to a state of affairs actually existing and which would continue to exist despite the coming to pass of what is supposed in the si-clause. Both of these things the speaker cannot express at one and the same time, though perhaps in some cases he finds it possible to follow a middle course by using the indicative when the verb chances to be modal. But with other verbs at any rate he must make a choice; by the use of the indicative he can assert the existing state of a/fairs, allow- ing the hearer to gather that the same state would continue under the adverse circumstances supposed, and on the other hand by employing the subjunctive he can confine himself to what woidd be true despite those circumstances, leaving it to the hearer to infer the actually existing state of affairs. Either mood is there- fore justified by the nature of the situation and the underlying thought. The indicative is the more vigorous and comprehensive form of expression • whereas the use of the subjunctive appeals to a mind trained to grammatical niceties as producing a more sym- metrical sentence structure. Consequently in Cicero the subjunctive is the normal and reg- ular usage ; e.g.^ p. Sulla 13.38 : Ne si argueret quidem turn denique ... id mihi crimino- sum videretur. r ii' ftf; 68 University of California Publications. [C'l-vss. Phil. When as here the supposition is contrary to fact, the choice of the secondary tenses of the subjunctive in the conclusion makes the speaker use the form of unreality of somethin*: which .is as a matter of fact true. Nevertheless in the orations alone there are some seventy cases in which a ^/-clause containinp< ratn us(|ue assiduo servos daf, si possiet meretricem illam invenire. Tri. 531-:]2: Em istuc oportet opseri mores malos si in opserendo possint inter fieri. It is noteworthy that in this group the verb of the .^/-clause is always quirr or posse. The thought of these clauses is akin to the purpose idea, but with a large admixture of doubt as to the attainment of the goal. A purpose clause with a parenthetical **if possible" or the like would in most cases provide a fair rendering for the thought; e.g. (Capt. 100-01), **He is buying up prisoners, that if possible he may light on one who can be exchanged for his son.'' In Cist. 184 ff. the .^/-clause appears to be an expansion of ei rci. Though the interpretation of such sentences is not difficult, it is hard to determine in a given case the precise shading of si. We can readily feel something of conditional force, as thouirh the word were chosen to convey uncertainty with regard to the attainment of the purpose. At the same time the English mind " Stutleniund's Studia, Vol. 1, p. 195. Vol. 1] Xutting.— Studies in the Si-clause. 77 is not slow here too to find the suggestion of interrogative mean- ing. For, in collocpual speech, with just such a virtual purpose idea to express we freely use the interrogative; e.g., *'I am going to the city (to see) if I can secure some tickets," i.e., *Ho secure some tickets if I can." The interrogative shading is most obtru- sive when the action of the main clause is a suggested experiment as in Tri. 531-32 ; there we may assume that other means of sup- pressing vicious practices have been tried, and the speaker now jocosely suggests that it would be well to make the experiment of planting them in that fatal field (to see?) if they too, as well as other things, will be killed off. (b). Sentences of other forms. The remaining cases of object ^/-clauses containing the sub- junctive and dependent on indicative forms are so few and scat- tering that they can be best presented under this general head. A.— Dependent on Verbs of Expectation and Waiting. Asin. 528-29 : An te id exspectare oportet, siquis promittat tibi te f acturum divitem, si moriatur mater sua ? Poen. 1391-92: lam pridem equidem istas scivi esse liberas et crspcctabam siqui eas assereret manu. Ps. 1148: lamdudum, si des, porrexi manum. Vid. 68 : Ilic astabo atque observabo, siquem amicum conspicer.^^ These sentences are manifestly like those cited of the form si sit — est. B.— Dependent on Verbs of Action and Effort. Capt. 27-28 : Coepit captivos commcrcari hie Aleos, siquem reperire posset, qui mutet suom (sc. filium). 32 Another example is probably to be found in True. 692-93, but it con- tains the ambiguous form opperiar. 78 University of California Publications. [Class. Phil. Merc. 622 ff. : Qiiin percontatu's hoiiiinis (|uae facies foret qui illani eniisset: eo si pacto posset indajrarier ninlier? Mil. 1207-08 : Nam si possem iillo niodo inipetrare, lit abiret nee te abdiiceret, (tprrnm dedi. Tri. 119-20: ei rei operam dare te fnrrnt aliqiianto aequius sifjiii probioreni facere posses. Vid. 56-57 : Ibo et quaeram, si(|uein possiin sociorum naneiseier sen qneni norini (pii advoeatus adsiet. Cf . Most. 837-38 : At tn isto ad vos optuere, quoniani eorniceni ne(inis eonspicari, si voltnrios forte possis eontni. Aniph. 880-81 : Mercnrinni iussi nie continuo vonseqiii, siqnid velleni iniperare. Mil. 1158: PA. Date niodo operam. AC. Id nos ad te, siquid velles, venimus. In this «;roup there is the same virtual purpose idea underlying? the **-elause that was found in sentences of the form si sit — est; and, as there, the verb of the .v/-elause is rejnilarly posse (quire), the exceptions bein^' the last two cases cited, which contain forms of velh,^^ Here too it is impossible to decide to what extent si is interrojrative. In Mil. 1207-08 however the unusual order (si- clause precedes) makes it difficult to feel any interro«>:ative force in si.^* In Tri. 119-20 ei rei ajjain anticipates the sZ-clause. Before attemptinjr to solve the problem of subjunctive ** pro- tasis" with indicative '^apodosis" for sentences containintr sub- junctive object clauses, it will be necessary to consider also those ^ There would be further exceptions if we shouKl inehnle Aul. 620-21 (persfrutabor, ai inveniam) and Pers. 44 (quaeram, siquis credat); these are excluded because of the presence of forms in -am, Cf. also Merc. 941, St. 151-52, the corrupt Tas. 806 and doubtful Amph. 621. '"Cf. Blase de mod. temp, permut. p. 22 (78). Lindskog (1. c. p. 73). without advancing any satisfactory evidence, is very decided in his disap- proval of Blase 's position. \ Vol. 1] Nutting.— Studies in the Si-clause, 79 cases in which an indicative object clause is used. The subjunc- tive examples were subdivided according to the nature of the verb of the main clause; (A) depending on verbs of expectation and waiting, (B) depending on verbs of action and effort. A similar plan will be followed here; but A is lacking, and it is necessary to add fCj— depending on verbs of seeing and know- ing. We therefore begin with B.— Dependent on Verbs of Action and Effort. Just as in the case of the subjunctive the verb of the s/-clause is here also regularly posse. Bacch. 1151 : Ego ad hunc iratum adgrcdiar, si possunuis nos hosce intro inlicere hue. Cist. 651-52: Ibo, perstquar iam ilium intro, ut haec ex me sciat eadem, si possum trancpiillum facere ex irato mihi. Cure. 701 : Aitimum advortitc hoc, si possum hoc inter vos componere. Men. 417-18 : adscntabor, quicquid dicet, mulieri, si possum hospitium nancisci. Men. 1048-49: Nunc ibo intro ad banc meretricem, (luamcpiam suscenset mihi, sei possum exorare ut pallam reddat. Rud. 890-91 : Verum tamen ibo, ei advoeatus ut siem, si(|ua mea opera citius — addici potest. Tri. 921: Quod ad exemplumst ? coniectura si reperire possumus. Tri. 958-59 : Enim vero ego nunc sycophantae huic sycophantari volo, si hunc possum illo mille nummum Philippum circum- ducere.^'^ In this group belong also a few conventionalized si vis clauses which find a parallel in two subjunctive examples already cited 38 Kud. 329 is doubtful in text and meaning. Cf. also Poen. 1063-64 and St. 740-41, which should perhaps come under this heading. oCl University of California Publications, [Class. Phil. II (si vellem, Arnph. 880-81, and si vclUs, ^Fil. 1158). All the eases here j^iveii perhaps do not contain object clauses, but the list is made complete so there may be no chance of excludinir what should be included. ♦ Aul. 209: litdeo ad te, Me«radore, sicjuid nie vis. Capt. 618: Do tibi operam, Aristophoiites, sicjuid est quod me velis. Men. i)HV): Em hie abiit, si vis per*se<|ui vesti*riis. Pers. 611 : Addiico banc, si((uid vis ex hac percontarier. Poen. 207-08: Em amores tuos, si vis spectare. Foen. 1047-48: Si itast, tesseram conferre si vis hospi- talem, eccam attuli. Tri. 51647: ST. Philto, te volo. PIT. Sicpiid vis, Stasime. As the sentences of this ji:roui) are compared with the corre- spondin«r examples with subjunctive .^/-clause, it nnist be con- fessed that one looks in vain for a difference of meaniiiir. It may be noted however that if a past tense is to be used in the si- clause the subjunctive is the mood chosen;''*' for all the indica- tive eases just cited employ the present tense. C— Dependent on Verbs of Seeing and Knowing. Bacch. 529: ibo ut risam hue ad eum, si fortest domi. Cas. 591 : Viso hue, amator si a foro rediit donnnu. Men. 142: lam si'iani, siquid titubatum.st, ubi reliquias videro. Mer. 155-56: Quin iam priusquam sum eloqutus scis, si mentiri volo. Pei-s. 825 : Vide vero, si tibi satis placet. =*Liiulskog (1. e. p. 69) nuikes this distinetion. Further he a«Uls (espe- cially with reference to poH.se and quire) that when there is a reference to the future, a verb in the first person takes the indicative and in other per- sons the subjunctive. 8o IJndsay, ('apt. 28 note. But Lindskog himself notices one exception to the latter part of the rule (namely Rud. 890-91), explaining it away by saying that mea opera addici potest is equal to poasum facere, ut addicatur. This is not altogether satisfying, especially as there is an exception to the other part of the rule which he does not notice, namely Vid. 56-57, where the first person subjunctive passim is used. Vol. 1] Nutting.— Studies in the Si-clause. 81 Tri. 748: Vid( si hoc utibile ma^ns atque in rem deputas. Tri. 763 : Sed vide consilium si placet.*^ In this group the nature of the verb of the main clause suggests most strongly interrogative force for si. Becker however (I.e. p. 195) holds that even here the word is not fully interrogative. For,Jie says, an undoubtedly interrogative word in some of the above cases woidd demand the subjunctive, according to Plau- tus' usage; here only the indicative is found. Treating only those cases which contain undoubted indica- tive and subjunctive forms, Plautus' usage in object clauses may be thus presented in tabular form. 1. After verbs of expectation and waiting the subjunctive is used. 2. After verbs of action and effort the mood varies. (a) In the present tenses both moods of posse are used; quire stands in the subjunctive, velle in the indicative. (b) In past tenses the subjunctive of posse and velle is employed. 3. After verbs of seeing and knowing the indicative is used. With the help of this outline it is possible by a process of exclu- sion to arrive at the probable cause of the use of the subjunctive in object clauses. For it may be remembered that in such clauses it was generally found to be true that the force of si was waver- ing between conditional and interrogative. The table just given shows that the use of the subjunctive must be due to the condi- tional force of the word— i.e., that this mood was chosen in accordance with the rule that called for it in regular conditional sentences. For in group 3 (after verbs of seeing and knowing), where the interrogative shading is most pronounced, the mood of the .9i-clause is always indicative. The weaker interrogative coloring of si in groups 1 and 2 cannot therefore have been the factor that caused the freciuent use of the subjunctive there. Cist. 682 is doubtful in meaning. 82 University of California Publications. [Class. Phil. Vol. 1] Nutting. — Studies in the Si-clause. 83 1*1 IV.-THE INDEFINITE SECOND SINGULAR. Baech. 440-41 : At nunc priuscniani soptnonnis est, si atfiufjas eiim inanii. extemplo puer paedagogo tabula disruinpit caput. Capt. 202: In re mala animo si bono ufarf\ adiuvat. Capt. 221 : Nam doli non doli sunt, nisi astu colas. Cas. 721 : Quia (piod teti^rere, ilico rapiunt: si eas ereptum, ilico scindunt. Ep. 674: Qua(|ue tan*rit, omne amburit. Si astes, aestu calefacit. Men. 108: Standumst in lecto, sicpiid de summo petas. Mil. 673 : Nam in mala uxore atque inimico siquid sumas, sumptus est. Fers. 449-50: Siquam rem ace u res sobrie aut frufjaliter solet ilia reete sum manus suecedere. Poen. 635-36: Malo siquid bene facias^ id beneficium interit. • Bono si(|uid male facias, aetatem expetit. Poen. 812-13: Siquid bene facias, levior plumast ^rratia. Siquid peecatumst, plumbeas iras «;erunt. Tri. 349: De majjnis divitiis siquid demas, plus fit an minus? Tri. 414-15 : Non tibi illud apparere, si sumas, potest, nisi tu immortale rere esse argentum tibi. Tri. 1053 : Si ma«re exigere occipias, duarum rerum exoritur optic. True. 461-62 : Nullani rem oportet dolose adsjrediri nisi astute adeurate(iue cxsequare. To these sentences of the form si sit — est apparently should be added one of the form si sit — erit: Amph. 703 ff . : Bacehae bacehanti si velis advorsarier, ex insana insaniorem facies, feriet saepius. Si ohscqnarc, una resolvas plaga. Though forms in -eris are strictly speaking: of uncertain mood, the two followinir eases may be at least enumerated in this connection : Poen. 212-13: Xain nullac maiiis res duae plus negoti hal)ent, forte si occcpcris exoruare. Tri. 1051 : Si(iuoi mutuom (juid dedcris, fit pro proprio perditum.^® A full and final explanation of the form of these sentences would naturally start with the subjunctive of the si-clause. But unfor- tunately the nature of this subjunctive is still a matter of uncer- tainty, and the material at hand is far too scanty to form the basis of any ade(|uate ctmclusion on that point. To reach such a conclusion it may be necessary to compass the wdde field in which the phenomenon of the concomitant relation between indefinite second siniiular and subjunctive mood manifests itself. How- ever, that there is a cause and effect relation involved cannot I think be for a moment doubted, the upholders of the other view notwithstandin*r. For so sweepin.s? is the tendency of a verb whose subject is the indefinite second sinjrular to go into the sub- junctive that Plautus offers but a single example of the form si (st—est to compare with the fourteen above of the form si sit — est : Asin. 241-42 : Portitorum simillumae sunt ianuae lenoniae: si ad fers, tum patent: si non est quod des, aedes non patent. Again a comparison of Poen. 812-13 and 635-36 (given above in full) is suggestive; in the second of these passages an alternative is afforded by siquid bene facias and siquid male facias, while in the other exactly the same thought finds expression in the clauses siepiid bene facias and siquid peccatumst. If the indefinite sec- ond singular has nothing to do with the use of the subjunctive •*Tri. 347-48 has a hortatory subjunctive in apodosis; cf. Aul. 380-81. 84 University of California Publications, [('lass. Phil Vol. 1] Nutting. — Studies in the Si-clause. 85 i| ' it is hard to account for the choice of iii(K)ds here. Thoutrh not in .sM'Uuist^s, the variation in mood is (juite as striking: in the two fonovvin«r cases: Mil. 947: V(»Iup est, (|nod at/as si id procedit lepide atrpie ex sententia. Pm'ii. 111)2: I 't vohip est honiini, niea soror, si cpiod ayit cluet victoria. '• Acceptinjr as a fact n()t yet satisfactorily cxphiincd the sub- junctive of the .s/-clause when the subject is the indefinite second sin«i:ular, the problem of subjunctive *' protasis" with indicative **apodosis" for the sentences under discussion is to determine why the subjunctive .s/-clause tlocs not level its conclusion. One looks in vain for a clear case of such level linjr in IMautus. There are it is true sentences like the following:: Cist. :iS : Eas si adeas, abitum <|uam aditum malis.*^ But tlie subject of the verb of the conclusion seems always to be as here the indefinite second sin«;ular, and the cases therefore give no pr(M)f of the workinjrs of a levelliny: force: for such a conclusion may take the subjunctive on its own merits, as is shown by exam[)les in which an indicative .s/-clause precedes: Bacch. 913 ff. : Lippi illic oculi st^rvos est simil limns: si non est, not is esse ne<|ue (hsidt res; si est, abstinere (pi in attinjjas non queas, Capt. llBtf.: Liber captivos avis ferae consimilis est : semel fujriendi si datast occasio, satis est — numquani postilla possis prendere. The reason why the subjiuictive si-clausc in the sentences under discussion does not level its conclusion is to be found in the nature of the underlyintr thou«rht. The .v/-clause refers to an action which the speaker assumes does happen, at least occasion- ally, and si is therefore practically a synonym of ubi or cmn*^ '"The 8hift in mood in the long passage Baeeh. 426 ff. and in Tri. 414-15 may have been caused by passing from the tiefinite to the indefinite second singular anr of cases in this cate(>ned to the end of the discussicm in order that it mi^ht not disti-act attention from more important matters if inserted in its lojrical place. I refer to the old problem of the difference in meaninjr of suppositions of the forms si sit and si erit. This (piestion is raised especially by what w^as said of the pure condi- tional sentence, namely that Plautus' failure to differentiate sharply between the uses of the two mood systems in general would be most likely to betray itself in the somewhat inter- changeal)le value of these two forms in particular, the time realm of bo'th beino- the future. That he did differentiate to a certain extent between the use of si sit and si erit is unquestionable, and I would suggest that the differentiation was partly on an objec- tive, partly on a subjective basis, i.e., that Plautus tenets to use the subjunctive in the two following cases : 40 Lindskog (1. c. p. 65) seems not to recognize Ps. 433 ff. as belonging to this category, thus leaving Cure. 265 as the only example of the use ot the subjunctive. To remove this exception to the rule he suggests that with B we read fit for sit. 88 University of California Publications. [Class. Phil. i (a) When there is actually less likelihood of fulfillinent. (b) When the speaker aims to give an impression of less like- lihood of fulfillment. The first of these cases is most strikingly illustrated by conces- sive sentences of the fonn si sit — erit. In at least three of the four examples found in Plautus the supposition of the .s"/-clause is extremely improbable. Amph. 450: Quadrigas si nunc inscendas lovis. Asin. 414: Siijuidem hercle nunc sumnium lovem te dicas detinuisse. Bacch. 1004: si iubeas maxime. Ep. 610: Si undecim deos praeter sese secum adducat luppiter. The peculiarity of these subjunctive cases may be brou«rht out into relief by contrastinjr the correspondinj; indicative exam- ples. Counting? as concessive one sentence in which the function of the A*t-clause is somewhat complicated, Plautus uses the form si crit — erit twice : Amph. 1048 ft*.: Ubi (piemque hominem aspexero si ancillani, sen servom, sive uxorem, sive adulteruni, seu patrem, sive avom videbo, obtruncabo in aedibus. Capt. 683-84: Si eiro hie peribo, ast ille ut dixit non redit, at erit mi hoc factum mortuo memorabile. In view of the fact that si est is a form that often has future force, there are doubtless some concessive sentences of the form si est — erit which should be added to the two of the form si erit — erit before making a comparison with the subjunctive cases first cited. I give the complete list, leaving it to the reader to choose those sentences in which the form si est seems to him to have future meaning. Whatever the sentences chosen it will still be clear that the sunjunctive tends to be used when the supposi- tion is extremely improbable, which is the point I am trying to illustrate. Asin. 405-06 : Siquidem hercle Aeacidinis minis animisque expletus ccdit, si med iratus tetigerit, iratus vapulabit. Vol. 1] Nutting.— Studies in the Si-clause. 89 Men. 1060-61 : Si voltis per oculos iurare, nilo hercle ea causa magis facietis ut ego hinc hodie abstulerim pallam. Most 229-30 : Siquidem hercle vendundust pater, venibit multo potius quam te . . . sinam egere. Kud. 1014 : Sei tu proreta isti navi's, ego gubernator ero. An illustration of Plautus' tendency to use the subjunctive when the speaker chooses to give an impression of unlikelihood of fulfillment is afforded by phrases of the form Quid si . . . sitf Not including the corrupt Cas. 806, there are nineteen ques- tions of this sort in Plautus. For the present purpose they may be subdivided according to person and number. (a). First person plural. Cas. 357-58 : Quid si propius attollamus signa eamusque obviam? Sequere. Cure. 303 : Quid si adeamusf hens, Curculio, te volo. Cure. 351 : Quid si abeamus, decumbamusf inquit. Consilium placet. Most. 393 : DEL. Quid si igitur abeamus hinc nos? TR. Non hoc longe, Delphi um. Poen. 330 : AG. Quid si adeamusf MI. Adeas. Poen. 707 ff.: Quid si evocemus hue foras Agorastoclem Ut ipsus testis sit sibi certissimus? Heus tu, qui furem captas, egredere ocius. Poen. 1162-63 : Quid si eamus illis obviam ? AG. At ne inter vias praeterbitamus metuo. Poen. 1249: HAN. Quid si eloquamurf AG. Censeo, hercle, patrue. 90 University of California Publications. [Class. Phil. Vol. 1] N Hit iny.— Studies in the Si-clause. 91 These (juestions corresixmd in *reneral to our '*Sii[)[)ose wv do thus and so'' — a form which leaves with the hearer the impres- sion that his wish or jud*rnient is beinjr consulted, and that the cominjr to pass of the thin«r su^'^rested is, frcmi the speaker's point of view, anythin*r but assured. Hut while in some of the above cases the hearer shows by his expression of a|)proval or disapproval that he feels himself consulted, in others (Cas. 857- 58, Cure. 80:^, and Poen. 707 ff.) the speaker really does not defer to his wish or jud»rnient at all, but without a i)ause pro- ceeds to do the thinir sujr^rested. That is, in certain cases the speak("r even thou^di he fully exjx'cts a thin*; to Ix* done, still uses in a somewhat perfunctory way a subjunctive phrase which appeal's to consult the wish or judjrment of the hearer. (b). First person singular. Capt. 612: HE. Quid ais? cpiid si adnim hunc insainim? TYX. Xufras: luditicabitur. Cist. 321 : Quid si adcam atipie apprUrm? Mali damnique inlecebra, salve. Cure. 145: PH. Quid si adrant ad fores at(|ue ocrrnffm? PA. Si lubet, necpie veto necpie iubeo. Ep. 543 : Quid si adcamf Pers. 724: TO. Quid si adtnoncam? VI. Tempus est. Poen. 728 : AG. Quid si recenti re aedis pultcmf ADV. Censeo. Rud. 535: CH. Quid si aliquo ad ludos me pro manduco loccmf LA. Quapropter? True. 6 : Quid si de vostro quippiam oremf — abnuont. With these may very properly be enumerated the single ease in which the perfect subjunctive is used : Capt. 599 : IIEG. fHercle cpiid si hunc comprehendi iusserim? TYN Sapias magis. In several of these cases the answer shows that the hearer feels that he is considted with regard to the speaker's action. There- fore the question in this number also is properly a form of defer- ential address. That however in some of the cases the speaker did not really mean to defer to the hearer's judgment is rendered probable by such an example as Cist. 321, where the ques- tion is spoken in solilocjuy and is practically an announcement of the speaker's intention— at any rate he at once proceeds to do the thing mentioned. Whenever this is true it provides another illustration of the use of the subjunctive to give the appearance of deferring to the hearer's judgment. Three cases remain which must be added to make the state- ment complete: (c). Third person singular. Bacch. 731-32 : MX. Quid scribam ? CH. Salutem tuo patri verbis tuis. PI. Quid si potius morbum, mortem scribal f id erit rectius. Merc. 419: Quid si igitur rcddatur illi unde emptast? True. 766 : 8ed (|uid ego hie clamo? quid si me iubeat intro mittier? In the first of these sentences scribal is clearly analogous to the first person use— the action proposed is put forward as a mere suggestion, here not by the actor himself but by another for him as it were. In the second case the verb is passivfe and the action devolves upon the first person; in meaning the sentence would properly be classed with those in which the subject of the verb is the first person. The third example is unique, and it seems to have nothing to do with the idiom under discussion aside from its likeness of form : its force is akin to that of indicative ques- tions of similar structure.*^ This completes the discussion of the difference of meanin of the forms si sit and si erit, and the paper might be closed at ♦• I omit from the eumeration Ps. 740 because it seems to have no exact parallel either among the subjunctive or the indicative cases. 1 have accepted the punctuation Quid? si opus sit ut dulce promat indidem, ecquid habet? and have treated the case as a pure conditional sentence. 4 i I University of Calif orniu Publications. [Class. Phil. this point. But havinjr ^ven the material in full for questions (»f the fonn Quid si . . . sitf I ou»rht perhaps to add for the sake of comparison those of the form Quid si . . . est (erit)? Because of stronjrly idiomatic use such a comparis(m throws veiy little direct lijrht on the question last under discussion (the dif- ference in jreneral between the meaninjr of si sit and si crit), but it is interestin*r in and for itself, and the matter seems to be nowhere fully treated.**^ The characteristic force of the indica- tive is seen most clearly in the following examples: Asin. 536-38 : CL. Non voto ted amare cpii dant, (pioia amcntur «rratia. PH. Quid si hie animus occupatusl, mater.' quid faciam? Merc. 890 : EV. Potin ut animo sis tran(|uiIlo? CHA. Quid si mi animus fluctuate Pers. 612-13: DO. Enim volo te adessc. TO. Ihni possum, quin hnic opci-juii dem hospiti (pioi erus iussit Quid si hie non volt me una adessc? Poen. 721-22: AG. Quid nunc mihi auctorcs cstis? ADV. I't frugi sis. AG. Quid si animus esse non siiiitf Rud. 1085-86 : TR. Xil i)eto nisi cistulam et crcpundia. (iR. Quid si ea sunt aureat Rud. 1138-39: Quid si ista aut superstitiosji aut hariola.^cf at((ue omnia quidquid inerit vera dicetf Tri. 1059-60: CH. Te volo. ST. Quid si ejro me te velle nolo? Questions like these arc not polite and deferential phrases. On the contrary they verge toward a protest ajrainst the expressed *" Liiulskog (1. C-. p. 106 if.) ^ives iiieoniplete lists. Of the subjiuu- tive cases he omits Mer. 419, Poen. Ili49, an«l Tnu*. 766; of the indica- tive eases, Amph. 7(»l, Baeih. 35, Pa. 286, Rud. 1()S6 and 1138 (two of these enijjioy the perfect tense which he does not treat at all) ; of cases of ambiguous form (-am), Rud. 1274 and 1312. Regarding the subjunctive he says (p. 109) **Rei natura fit, ut semper praesens coniunetivi usurpe- tur;" but C'apt. 599 has iusserim. O. Brugmann (1. c. p. 27) touches on this subject, but with very incomplete material. Of. Brix on Capt. 613 and Sounenschein on Bud. 472. Vol. 1] Nutting. — Studies in the Si-clause. 93 desire or advice of the person addressed. In translation we instinctively recotrnize this fact by bejxinninjr with an adversa- tive pai-ticle ''But what if . . . V The tone of the question mav be even insolent, as in the last case cited.*" Other examples of a similar nature but with the speaker's feelin«r of protest or hesitation perhaps not so clearly marked are : Amph. 391-92 : SO. Tuae fidei credo : .AIE. Meae. SO. Quid si fallrsf Asin. 193 IT. : Si mihi dantur duo talenta ar*renti niunerata in manum, banc tibi noctem honoris causa *:ratiis dono dabo. AR. Quid si non cstf Bacch. 1184-85: NT. Quem iguous forms in -am, namely Amph. 313, Merc 504, 578, Most. 1093, Rud. 1274, and 1311 ff.; all excepting the last have subjunctive force. Two cases have verbs terminating in -erit (Cas. 345 and Rud. 472 ff.) and one with the form fajris (Mil. 1417) ; these three have indicative force. F'iil 'Pi^ UNIVERSITY OF CALIFORNIA PUBLICATIONS-(CONTINUED) ASTRONOMY.-W. W. Campbell, Editor. Publications of the Lick Observatory.— Volumes I-V completed. Volume VI (in progress): No. I. A Short Method of Determining Orbits from Three Observations, by A. O. Leuschner. No. 2. Elements of Asteroid 1900 GA, by A. O. Leuschner and Adelaide M. Hobe. No. 3. Preliminary Elements of Comet 1900 III, by R. H. Curtiss and C. G. Dall. Contributions from the Lick Observatory.— Nos. 1-V. Lick Observatory Bulletins.— Volume I (pp. 193) completed. Volume II (in progress). BOTANY.-W. A. Setchell, Editor. Price per volume $3.50. Volume I (pp. 418) completed. Volume II (in progress): No. I. A Review of Californian Polemoniaceae, by Jessie Milliken. Price, $0.75 No. 2. Contributions to Cytological Technique, by W. J.V. Osterhout. Price, .50 EDUCATION.— Elmer E. Brown, Editor. Price per volume $2.50. Volume I (pp. 424). Notes on the Development of a Child, by Milicent W. Shinn Price, 2.25 Vol. II (in progress). — No. 1. Notes on Children's Drawings, by Elmer E. Brown Price, .50 Vol. Ill (in progress). —No. 1. Origin of American State Universities, by Elmer E. Brown Price, .50 No. 2. State Aid to Secondary Schools, by David Rhys Jones Price, .75 GEOLOGY.— Bulletin of the Department of Geology. Andrew C. Lawson, Editor. Price per volume $3.50. Volumes I (pp. 428), II (pp. 450) and III (475), completed. Volume IV (in progress): No. 1. The Geology of the Upper Region of the Main Walker River, Nevada, by T. D. Smith (in press). PATHOLOGY.— Alonzo Englebert Taylor, Editor. Price per volume $2.00 Volume I (in progress): No. I . On the Quantitative Separation of the Globulins of Hemolytic Serum, with Special Reference to the Carbon Dioxide Group, by Clarence Quinan. No. 2. Hydrolysis of Protamine with Especial Reference to the Action of Trypsin, by Alonzo Englebert Taylor. No. 3. On the Synthesis of Fat Through the Reversed Action of a Fat- Splitting Enzyme, by Alonzo Englebert Taylor. No. 4. On the Occurrence of Amido-Acids in Degenerated Tissues, by Alonzo Englebert Taylor. No. 5. On the Autolysis of Protein, by Alonzo Englebert Taylor. No. 6. On the Reversion of Tryptic Digestion, by Alonzo Englebert Taylor No. 7. Studies on an Ash-Free Diet, by Alonzo Englebert Taylor. In one cover. ) In > one •J cover. VNIVERSII'Y Hr CALIFORNIA PUBLICATIONS-(CONTINUCO) 'XBiiiiisoF: MJL mk ■♦ "•"IT'' Volume I| completed. Price, $2.00 Ho. .Mo. Nol 9* No. 7. No. 1. Tlie Siiiiimiim Bonuni, by Evander Bradley McGilvary. 2. The Essentials of Human Faculty, by Sidney Edward Mezes. 3. Some Scientific Apologies for Evil, by George Malcolm Stratton. Pragmatism and the a priori^ by Charles Henry Rieder. Latter-Day Flowing-Philosophy, by Charles Montague Bakewell. Some Problems in Evolution and Education, by Ernest Norton Henderson. Fiilosophy and Science in tii Stpdy of Education, by Jesse Dismukes Burks. 1. The Dialectic of Bruno and Spinoza, by Arthur Oncken Lovejoy. 0. The Logic of Self-Realization, by Henry Waldgrave Stuart. 10. Utility and the Accepted Type, by Theodore de Lopez de Laguna. U. A Theory of the Syllogism, by Knight Dunlap. 12. The ^s^&Skl Principle of Tri|tJi-E¥iilMi|i#n, by Harry Allen Overstreet. ■'11 «■ II' i " " !' In one cover. PHYSIOLOOT.—Jacques Loeb, Editor. Price per volume $2.00. Volume I (pp. 217) completed. Volume II (in progress): No. 1. The Control of Heliotropic Reactions in Fresh Water Crustaceans by Chemicals, Especially Cog (a preliminary communication), by Jacques Loeb. Mil. 2. further E^cperiments on Heterogeneous Hybridization in Echino- derms, by Jacques Loeb. No. 3. Tlic Influence of Calcium and Barium on the Secretory Activity of the Kidney (second communication), by John Bruce Mac- Callum. No. 4. lote on the Galvanotropic Reactions of the Medusa Polyorchis Penicillata A. Agassiz, by Frank W. Bancroft. Ho. 5, Tlie Action on the Intestine of Solutions Containing Two Salts, 1 , by John Bruce MacCallttm '" I one No, f. The Action of Purgatives in a Crustacean {Sida Crysiallina) , by cover John Bruce MacCallum. J Z00L0(3lf.^W. 1. Ritter, Editor. Price per volume $3.50. Volume (pp. 286) completed. Volume II (in progress): No. 1. The Hydroids of the San Diego Region, by Harry Beal Torrey. 1 In Pages 43, text figures 23. I one lit.i fie Ctenophores of the San Diego Region, by Harry Beal | PHce Torrey. Pages 6, Plate 1 . J .60 UHmmSlfY CH»OinCLE.-An official record of University life, issued quarterly, edited by a committee of the faculty. Price, $1.CX) per year. Current volume No. VII. Address all orders, or requests for information concerning the above publications (except Agricultural) to Tlie University Press, Berkeley, California.