^■<: '■<--'^'*^'.<;,/»5!v' *, ''H-^^>: By/ (fJorncU Untttsrattg ffiiihrarg Cornell University Library PA 2297.B81 The ndlcatlve indirect question in L^ti 3 1924 021 617 588 Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924021617588 aHjr Ittwf raitg of Qlijtrago THE INDrCATIVE INDIRECT QUESTION IN LATIN A DISSERTATION SIJBMITTEIX TO THE FACULTY OF THE GRADUATE SCHOOI, OF ART9~AND LITERATURE IN CANDIDACY FOR-THE DEGREE OF doctor" OF PHiLOS'OPHY DEPARTMENT OF LATIN , BY ALICE FREDA BRAUNLICH Private Edition, Distributed By THE UNIVERSITY OF CHICAGO LIBRARIES CHICAGO, ILLINOIS, 1920 Ql^f lltttoeraitg at OII;tra$o THE INDICATIVE INDIRECT QUESTION IN LATIN A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF ARTS AND LITERATURE IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF LATIN BY ALICE FREDA BRAUNLICH Private Edition, Distributed By THE UNIVERSITY OF CHICAGO UBRARIES CHICAGO, ILLINOIS 1920 PREFACE The writing of this dissertation was suggested to me by Professor Hale and has been carried out under his direction. With the title, "The Indicative Indirect Question in the Latin of the Repubhcan Period, " my doctoral dissertation was completed in 1913. At Professor Hale's suggestion, it was subsequently enlarged to its present scope. To Professor Hale I am indebted, not only for the subject and plan of this paper, but also for generous help in the execution of the plan, for many of the examples from Latin authors, for terminology, and for methods of work. While he is not to be held responsible for every detail of the paper, yet he has carefully guided my investigation throughout its course and has read and criticized the entire dissertation, both in its earliest draft and in its all-but-final form. I acknowledge my indebtedness for help and encouragement in research, also to aU the other members of the departments of Latin, Greek, and Comparative Philology, of the Univiersity of Chicago, and especially to Professor Beeson, who read and criticized this entire paper in one of the early drafts and has since made numerous valuable suggestions, and to Professor Prescott, who has generously aided me in the textual study of many of the citations. m TABLE OF CONTENTS PREFACE iii BIBLIOGRAPHY k INTRODUCTION I. Review of Various Studies and Opinions Concerning the Indicative Indirect Question xvii II. Purpose and Method of This Study xxviii PART I. A STUDY OF THE CONSTRUCTION TO THE END OF THE AUGUSTAN AGE Chapter I. An Attempt at Defining the Direct and the Indbrect Question 1 Chapter II. Indeterminate Examples: Indirect or Direct Questions I. Questions Whose Interpretation is IndiflEerent 6 n. Questions which if an Indicative Indirect Question Exists, are more Nat- urally Interpreted as Indirect, rather than as Direct. 1. Introductory 9 2. Indications that Questions are Indirect, Rather than Direct and not Quoted 10 3. Examples of Questions More Naturally Interpreted as Indirect, than as Direct and Not Quoted 12 4. Indications that Questions are Indirect, Rather than Directly Quoted 24 5. Examples of Questions More Naturally Interpreted as Indirect than as Directly Quoted 26 Chapter HI. Indeterminate Examples: Indirect Questions or Exclamations I. Clauses Whose Interpretation is Indifferent 29 II. Clauses Which, if an Indicative Indirect Question Exists, are More Naturally Interpreted as Indirect Questions than as Exclamations 1. Indications that Clauses are Indirect Questions Rather than Exclama- tions 31 2. Examples of Clauses More Naturally Interpreted as Indirect Questions than as Exclamations 33 V vi The Indicative Indirect Question in Latin Chapter IV. Indeterminate Examples: Indirect Questions or Relative Clauses I. Clauses Whose Interpretation is Indifferent 1. Introductory ^ 2. Examples of Clauses Whose Interpretation is Indifierent 45 II. Clauses Which, if an Indicative Indirect Question Exists, are More Naturally Interpreted as Indirect Questions than as Relative Clauses 1. Indications that Clauses are Indirect Questions Rather than Relative Clauses 52 2. Examples of Clauses More Naturally Interpreted as Indirect Ques- tions than as Relative Clauses 56 Chapter V. Indeterminate Examples: Indirect Questions or Conditions I. Introductory: Review of Opinions Concerning the Interrogative Use of "Si" 75 II. Clauses Whose Interpretation is Indifierent 1. Introductory 76 2. Examples of Clauses Whose Interpretation is Indifferent 78 III. Clauses which, if an Indicative Indirect Question Exists, are More Naturally Interpreted as Indirect Questions than as Conditions 1. Indications that Clauses are Indirect Questions Rather than Condi- tions 79 2. Examples of Clauses More Naturally Interpreted as Indirect Ques- tions than as Conditions 80 Chapter VI. Indeterminate Examples: Miscellaneous 85 Chapter VII. The Proofs that Clauses are Indirect Questions I. The Discrimination of the Indirect Question from All Other Kinds of Clauses 87 II. The Discrimination of the Indirect from the Direct Question 87 III. The Discrimination of the Indirect Question from the Exclamation . . 91 IV. The Discrimination of the Indirect Question from the Relative Clause 92 V. The Discrimination of the Indirect Question from the Condition. ... 94 Chapter VIII. Clauses Which, it the Sole or Best Manuscript Tradi- tion is Followed, are Certainly Indicative Indirect Questions 96 Chapter DC. Preliminary Conclusion 105 Chapter X. The Disposal op the Indicative Indirect Question by Emendation 106 Contents vii PART n. A SUPPLEMENTARY STUDY OF THE CONSTRUCTION IN LATER LATIN Chapter I. Indeterminate Exampi^s: Indirect or Direct Qdestions I. Questions Whose Interpretation is Indifferent 110 II. Questions Which are More Naturally Interpreted as Indirect than as Direct 1. Questions More Naturally Interpreted as Indirect, than as Direct and Not Quoted Ill 2. Questions More Naturally Interpreted as Indirect, than as Directly Quoted lis Chapter II. Indeterminate Examples: Indirect Questions or Exclamations I. Clauses Whose Interpretation is Indifferent 118 II. Clauses Which are More Naturally Interpreted as Indirect Questiohs than as Exclamations 118 Chapter III. Indeterminate Examples: Indirect Questions or Relative Clauses I. Clauses Whose Interpretation is Indifferent 125 n. Clauses Which are More Naturally Interpreted as Indirect Questions than as Relative Clauses 127 Chapter IV. Indeterminate Examples: Indirect Questions or Conditions I. Clauses Whose Interpretation is Indifferent 139 II. Clauses Which are More Naturally Interpreted as Indirect Questions than as Conditions 139 Chapter V. Indeterminate Examples: Miscellaneous 142 Chapter VI. Clauses Which, if the Sole or Best Manuscript Tradi- tion IS Followed, are Certainly Indicative Indirect Questions 144 Chapter VII. Preliminary Conclusion 158 GENERAL CONCLUSION I. The Occurrence of Instances of the Indicative Indirect Question 158 n. Unjustifiability of Emendation and Forced Interpretation 158 III. Comparative Frequency of the Construction in Conversational and In- elegant Styles 160 IV. Outline of the History of the Construction 164 V. Comparative Frequency of the Construction in Early and in Late Latin 165 viii The Indicative Indirect Question in Latin VI. The Construction Seldom Used as an Archaism 166 Vn. The Construction Not a Grecism 167 Vni. The Influence of Meter 168 IX. Avoidance of Ambiguity, by the Use of the Indicative 169 X. The Subjunctive the More Usual Mood in the Indirect Question of Fact 169 XI. Limitations of the Use of the Indicative 170 Xn. Parallel Use of Indicative and Subjunctive 171 Xin. Application of the Results of this Paper to the Study of the Romance Languages 173 XIV. Application of the Results of this Paper to the Study of Latin 175 APPENDIX I Additional Remarks on the Mood Use in Indibect Qttestions in Late Latin 176 APPENDIX II Supplementary List of Doubtful Passages, in Most of Which the Evi- dence Favoes the Subjunctive Reading 182 Manifestly Wrong Examples of the Indicative Indirect Question .... 190 APPENDIX III Relative (Quis) Quid I. Introductory 191 II. Evidence for the Use of (Quis) Quid, in the Sense of (Quisquis) Quicquid 193 III. Evidence for the Use of (Quis) Quid in the Sense of (Qui) Quod 196 IV. Conclusion 199 APPENDIX IV The Intekrogative Use of "Si" 201 INDEX LOCORUM 205 BIBLIOGRAPHY I. Latin Authors The Latin authors referred to in this study, exclusive of the appendices, and the editions chiefly used are mentioned here. *Ad Heeenniom: F. Marx. Leipzig, 1894. C. F. W. MiiUer. Leipzig, 1893. ♦Aetheria (or Silvia) : E. A. Bechtel. Chicago, 1902. Ammianus Marcelldtos: C. U. Clark, Vol. I, Libri 14-25. Berlin, 1910. *ApuiEros: R. Helm and P. Thomas. Leipzig, 1905-10. *AscoNias: A. Kiessling and R. Schoell. Berlin, 1875. A. C.Clark. Oxford, 1907. Augustine: B. Dombart, De Civitate Dei, ed. 2. Leipzig, 1877-92. P. Knoll, Confessiones. Leipzig, 1898. J. P. Migne, Patrologia Latina, vols. 32-47. *Caesar: B. Kubler, Editio Maior. Leipzig, 1893-97. H. Meusel, Lexicon Caesarianimi. Berlin, 1887. *Cato: H. Keil, De Agri Cultura. Leipzig, 1884. H. Jordan, Praeter Librum de Re Rustica Quae Extant. Leipzig, 1860. *Catui,lus: R. Ellis. Oxford, 1904. Celsus: C. Daremberg. Leipzig, 1891. •Cicero: C. F. W. Miiller. Leipzig, 1893-98. J. G. Baiter and C. L. Kayser. Leipzig, 1860-69. C. Halm and J. G. Baiter, ex Recensione Orelli. Zurich, 1845-61. J. A. Ernest. London, 1819. A. C. Clark, Orationes pro P. Quinctio, pro L. Flacco, etc. Oxford, 1909. R. Y. Tyrrell and L. C. Purser, The Correspondence of Cicero. Dublin, 1885-99. *The asterisk indicates that the author or the work that is starred has been read completely through, for the purposes of this study. ix X Bibliography L. Mendelssohn, Epistularum Libri Sedecim. Leipzig, 1893. L. C. Purser, Ad Atticum. Oxford, 1903. F. Hofmann and G. Andresen, Ausgewahlte Briefe, ed. 2. Berlin, 1885. A. S. Wilkins, Rhetorical Works. Oxford, 1901-03. Claudiants Mamertius: A. Engelbrecht. Vienna, 1885. *Coltjmella: . J. G. Schneider, Scriptores Rei Rusticae Veteres Latini. Leipzig, 1794-97. Commodian: E. Ludwig. Leipzig, 1877-78. *CoEippxJS: M. Petschenig, Berliner Siudienfiir hlassische Philologie IV (1886). *Enni0s: J. Vahlen. Leipzig, 1903. Ennodhjs: A. Dubois, La Latinitfi d' Ennodius. Paris, 1903. Festtjs: W. M. Lindsay. Leipzig, 1913. Pronto: S. A. Naber, M. Comelii Frontonis et M. Aurelii Imperatoris Epistulae. Leipzig, 1867. Gellius: C. Hosius, Volumen Prius (Libri 1-10). Leipzig, 1903. M. Hertz. Berlin, 1883. *GREGORy 01' TotJRS: W. Amdt and B. Krusch, in Monumenta Germaniae Historica. Hanover, 1885. *Hermerus, MuLoiTEDicnsiA Chironis: E. Oder. Leipzig, 1901. "Horace: F. Vollmer. Leipzig, 1907. Sermones A. Palmer, ed. 4. London, 1891. P. Lejay- Paris, 1911. L. Muller. Vienna, 1891. EpisUdae P. Lejay, in edition of Horace by F. Plessis and P. Lejay. Paris, 1912. Jerome: J. P. Migne, vols. 22-29. *Juvenal: A. E. Housman. London, 1905. *Livy: W. Weissenbom. Leipzig, 1894. R. S. Conway and C. F. Walters, Books 1-S. Oxford, 1914. *Lucan: C. Hosius. Leipzig, 1905. Bibliography XI *LuciLros: F. Marx. Leipzig, 1904-05. *LirciLius (?), Aetna: S. Sudhaus. Leipzig, 1898. R. EUis, in Postgate, Corpus Poetarum Latinorum Fasc. III. London, 1900. H. A. J. Munro. Cambridge, 1867. *LTTCEETros: H. A. J. Munro. London, 1900. •Lygdamus: In Tibullus, g.v. MANiuns: T. Breiter. Leipzig, 1908. J. van Wageningen. Leipzig, 1915. H. W. Ganod, Liber II. Oxford, 1911. ♦Maeiiai: L. Friedlander. Leipzig, 1886. W. M. Lindsay. Oxford, 1902. *NAEvins (no examples): L. Muller, Enni Carminum Reliquiae. Accedunt Cn. Naevi belli Poenici quae supersunt. St. Petersburg, 1884. Nepos: E. O. Winstedt. Oxford, 1904. *Ovid: M. Haupt, O. Kom, and R. Ehwald, Metamorphoses, ed. 3. Berlin, 1898. H. Magnus, Metamorphoses, ed. 1 and 2. Gotha, 1886-1903; also ed. Berlin, 1914. R. Ehwald, Vol. I, Amores, Epist., Medic. Fac. Fem., Ars Am., Rem. Am., Leipzig, 1888; Vol. II, Metamorphoses, Editio Maior., Leipzig, 1915. P. Brandt, Amores. Leipzig, 1911. Ars Amatoria. Leipzig, 1902. A. Pahner, Heroides. Oxford, 1898. O. Giithling, Fasti, Tristia, Ibis, Ex Ponto, HaJieutica, Fragmenta. Prague, 1884. R. Merkel, Tristia, Ibis, Ex Ponto, Fasti. Leipzig, 1897. S. G. Owen, Tristia, Ex Ponto, Halieutica, Fragm. Oxford, 1915. R. Ellis, Ibis. Oxford, 1881. *PEESrus: O. Jahn, Persii, luvenalis, Sulpiciae Saturae. Berlin, 1910. *PETEONros: F. Biicheler, Saturae et Liber Priapeorum. Berlin, 1862. F. Biicheler— W. Heraeus. Berlin, 1912. L. Friedlander, Cena Trimalchionis. Leipzig, 1891. *Plautus: F. Ritschl, G. Loewe, G. Goetz, F. Schoell. Leipzig, 1878-94. G. Goetz— F. Schoell. Leipzig, 1904-09. F. Leo. Berlin, 1895-96. W. M. Lindsay. Oxford, 1903. Xii Bibliography J. L. Ussing. Hafnia, 1875-86. C. H. Weise. Quedlenburg and Leipzig, 1848; index at end of Vol. II. E. P. Moms, Pseudolus. Boston, 1895. E. A. Sonnenschein, Rudens. Oxford, 1891. E. A. Sonnenschein, Mostellaria. Oxford, 1907. A. R. S. Hallidie, Captivi. London, 1891. E. P. Morris, Captivi and Trinummus. Boston, 1898. Pliny the Elder: L. Janus, Vol. I. Leipzig, 1870. *Pliny the Yotjnger: C. F. W. Miiller. Leipzig, 1903. R. C. Kukula. Leipzig, 1908; 2d ed. 1912. Priscian: H. Keil, Grammatici Latini, Vols. II and III. Leipzig, 1855-59. *PROPERTnJS: C. Hosius. Leipzig, 1911. J. S. Phillimore. Oxford, 1901. J. P. Postgate, Select Elegies of Propertius. London, 1897. *QuiNTn,LW, Instiiuiio Oratoeia: L. Radermacher, Part I (Books I-VI). Leipzig, 1907. F. Meister (XII Books). Leipzig, 1886-87. RuTiLius LupTJS: In Halm, Rhetores Latini Minores, q. v. *Salltjst: R. Dietsch. Leipzig, 1859. •Seneca the Elder: F. Haase. Leipzig, 1902. *Seneca the Younger: C. R. Fickert, Epistulae Morales, Leipzig, 1842; De Beneficiis, De Clementia, De Ira, Leipzig, 1843. F. Haase, Opera quae Supersunt, Leipzig, 1887-92; Opera quae Supersunt Sup- plementum, Leipzig, 1902. J. Vahlen, Dialogi. Ulm, 1884. C. Hosius, De Beneficiis and De Clementia. Leipzig, 1914. A. Gercke, Naturales Quaestiones. Leipzig, 1907. R. Peiper and G. Richter, Tragoediae. Leipzig, 1902. F. Leo, Tragoediae. Berlin, 1878. *SlLros Italicus: L. Bauer. Leipzig, 1890. •Statius: A. Klotz, Leipzig: Silvae, 1900; Achilleis, 1902; Thebais, 1908. Suupicros Severxjs H. Goelzer, Granmiaticae in Sulpicium Severum Observationes. Paris, 1883. Tacitus: C. D. Fisher, Annales, Oxford, 1906; Historiae, Oxford, 1910. H. Fumeaux, Germania. Oxford, 1894. Bibliography xiii *Terence: R. Y. TyrreU. Oxford, 1902. A. Fleckeisen. Leipzig, 1895. K. Dziatzko — E. Hauler, Phormio. Leipzig, 1898. H. C. Elmer, Phormio. Boston, 1895 and 1916. G. Laing, Phormio. Chicago, 1908. P. Thomas, Hecyra. Paris, 1887. Tertullian: F. Oehler. Leipzig, 1853-54. *TiBERiAiros: E. Baehrens, Unedierte lateinische Gedichte. Leipzig, 1877. *TiBULLtrs: J. P. Postgate. Oxford, 1914. Valerius Flaccus: O. Kramer. Leipzig, 1913. Valerius Maximus: C. Kempf, ed. 2. Leipzig, 1888. C. B. Hase. Paris, 1822-23. *Vaiiro: H. Keil, Res Rusticae. Leipzig, 1884. G. Goetz, Res Rusticae. Leipzig, 1912. G. Goetz and F. Schoell, De Lingua Latina. Leipzig, 1910. A. Riese, Saturae Menippeae. Leipzig, 1865. *Vegetius: J. G. Schneider. Leipzig, 1797. E. Lommatzsch. Leipzig, 1903. *Virgil: O. Guthling. Leipzig, 1894-1907. F. A. Hiitzel. Oxford, 1900. A. Sidgwick. Cambridge, 1905; index at end of Vol. II. E. Norden, Aeneid, Buch VI. Leipzig, 1903. F. Leo. Culex, Accedit Copa Elegia. Berlin, 1891. T. Ladewig, C. Schaper, and P. Deuticke, erstes Bandchen, Bukolika und Geor- gika, neunte Auflage, bearbeitet von P. Jahn. Berlin, 1915. ViTRUvius: F. Krohn. Leipzig, 1912. II. Modern Collections *Bruns, C. G. and Gradenwitz, O. Fontes luris Romani Antiqui, Pars Prior. Tubingen, 1909. *Buecheler, F. Cannina Latina Epigraphica, Leipzig, 1895-97; in Buecheler-Riese, Anthologia Latina. Corpus Inscriftionuh Lahnarum. Consilio et Auctoritate Academiae Litterarum Regiae Borussicae. Berlin, 1893-99. xiv Bibliography Halm, K. Rhetores Latini Minores. Leipzig, 1863. HUEBNER, E. Inscriptiones Hispaniae Christianae. Berlin, 1871. *Meyer, H. Oratorum Romanorum Fragmenta. , Zurich, 1842. MiGNE, J. P. Patxologia Latina. Paris, 1878-90. Peter, H. ♦Veterum Historicorum Romanorum Reliquiae, Vol. I. Leipzig, 1870. Historicorum Romanorum Reliquiae, Vol. I. Leipzig, 1914. *Scriptores Historiae Augustae. Leipzig, 1884. RlBBECK, O. Scaenicae Romanorum Poesis Fragmenta. Leipzig, 1871-73. III. Modern Books, Articles, and Reviews Only the works that have been chiefly used are cited here. Some additional works are listed in Appendix I. Appel, E. Beitrage zur Erklarung des Corippus. Munich, 1904. Baehrens, W. a. Beitrage zur lateinischen Sjmtax, Philologus SuppUmentband XII (1912) 235-556. Becker, E. De Syntaxi Interrogationum Obliquarum apud Priscos Scriptores Latinos, in Studemund, Sfudien auf dem Gebiete des archaischen Lateins, Vol. I, pp. 113- 314. Berlin, 1873. Bennett, C. E. Syntax of Early Latin, Vol. I, The Verb. Boston, 1910. Bonnet, M. Le latin de Gr6goire de Tours. Paris, 1890. Brugmann, K. Kurze vergleichende Grammatik der indogermanischen Sprachen. Strassburg, 1904. Buck, C. D. A Grammar of Oscan and Umbrian. Boston, 1904 (German translation, Heidelberg, 1905). Deecke, G. De Usu Pronominls Relativi apud Poetas Veteres Latinos Quaestiones Syntacticae. Gottmgen, 1907. Delbruck, B. Vergleichende Sjmtax der indogermanischen Sprachen, Part III. (Brugmann- Delbriick, Grundriss der vergleichenden Grammatik der indogermanischen Sprachen, Vol. V). Strassburg, 1900. Draeger, a. Historische Syntax der lateinischen Sprache, Vol. II. Leipzig, 1881. Bibliography xv Gapfiot, F. Quelques cas d' interrogation indirecte, Revue de philologie XXVIII (1904) 49-SS. Ecqui fuerit si particulae in interrogando latine usus. Paris, 1904. Le prgtendu emploi de si interrogatif en latin, Reiiue de philologie XXXII (1908) 47-58. Pour le vrai latin. Paris, 1909. (Quis) Quid relatif, Revue de philologie XXXIV (1910) 57-67. GUTSCHE, W. O. De Interrogationibus Obliquis apud Ciceronem Observationes Selectae. Halle 1885. Hale, W. G. The Cum-Constructions: their History and Functions, Cornell Stiidies in Classical Philology I. Ithaca, N. Y., 1887 (German translation by A. Neitzert. Leip- zig, 1891). A Century of Metaphysical Syntax, Congress of Arts and Science, Universal Exposition, Vol. Ill, pp. 191-202. St. Louis, 1904. Hale, W. G. and Buck, C. D. Latin Grammar. Boston, 1903. Keoll, W. Der kteinische Relativsatz, Glotta III (1910-12) pp. 1-18. KtJHNER, R. Ausfuhrliche Grammatik der lateinischen Sprache, Zweiter Band, Satzlehre, neubearbeitet von C. Stegmann. Hanover, 1914. Langen, p. Commentationes Comificianae, Philologus XXXVII (1877) pp. 385-414. Lindsay, W. M. Syntax of Plautus. Oxford, 1907. LiNDSKOG, C. De Enuntiatis apud Plautum et Terentium Condicionalibus. Lund, 1895. Quaestiones de Parataxi et Hypotaxi apud Priscos Latinos. Lund, 1896. Maex, F. Die Beziehimgen des Altlateins zum Spatlatein, Neue Jahrbucher fUr das klass- ische Altertum XXIII (1909) pp. 434-448. Meyer-Ltjbke, W. Grammaire des langues romanes, traduction franfaise par Doutrepont; tome 3, syntaxe. Paris, 1900. Die vorromanischen Volkssprachen, in Grober, Grundriss der romanischen Philo- logie Vol. I. Strassburg, 1888. MOERIS, E. P. On the Sentence Question in Plautus and Terence, The American Journal of Philology X (1889) pp. 397-436 and XI (1890) pp. 16 ff. and 145 ff. Owen, E. T. Interrogative Thought and the Means of its Expression, Transactions of the Wis- consin, Academy of Sciences, Arts, and Letters, XIV. Madison, Wis., 1903, pp. 345-470. xvi Bibliography Paetzolt, F. De Latini Pronominis Relativi Syntaxi Prisca. Breslau, 1873. PiRSON, J. La langue des inscriptions latines de la Gatde. Brussells, 1901. Pratjn, J. Bemerkungen zur Syntax des Vitruvs. Bamburg, 1885. Prescott, H. W. Review of Helm's edition of Apuleius, Metamorphoses, Classical Philology X (1915) 358. Patjl, H. Prinzipien der Sprachgeschichte, ed. 3. HaUe, 1898. Rebldjg, O. Versuch einer Charakteristik der romischen Umgangssprache. Kiel, 1873. RiEMANN, O. La langue et la grammaire de Tite-Live. Paris, 1885. RiEMANN, O., and Goelzer, H. Grammaire comparfie du grec et du latin. Paris, 1897. SCHMALZ, J. H. Lateinische Syntax, in Stolz-Schmalz, Lateinische Grammatik, ed. 4. Mimich, 1910. Skutsch, F. Review of Gaffiot, Pour le vrai latin, Glotta III (1912) 366. Strobel, E. Tulliana. Munich, 1908. Uhlmann, W. De Sexti Properti Genere Dicendi. Munster, 1909. WOLFrLIN, E. Bemerkungen uber das Vulgariatein, Philologus XXXIV (1876) 137-165. Wolff, O. De Enuntiatis Interrogativis apud Catullum, Tibullum, Propertium. Halle, 1883. INTRODUCTION I. Review of Various Studies and Opinions concerning the Indicative Indirect Question The subject of the use in Latin of the indicative indirect question is one which has been much discussed but upon which there is still no gen- eral agreement. The study of the problem has been hampered by various preconceived ideas, which once prevailed, and still have influence, in the field of syntax. According to a theory which originated in a careless naming of one of the moods by the Greek grammarians, and the translating of the resulting Greek name into Latin, the subjunctive is the mood of dependence.* This theory has led some scholars to maintain that all dependent questions should be in the subjunctive, and that apparent examples of such questions with the indicative are not really dependent questions, biit are to be otherwise interpreted.^ This theory has led, likewise, to the explanation of the use of the indicative mood in questions apparentiy indirect, as due to the tendency of popular speech to disregard the relationships of clauses.' According to another theory, the indicative is the mood of objective reality, whUe the subjunctive is the mood of thought.* Therefore, the ' See Hale, "A Century of Metaphysical Syntax," Congress of Arts and Science, Universal Exposition (St. Louis, 1904), Vol. Ill, 191 flf., especially 195 and 197. For the refutation of this theory, cf. Hale, The Cum-Constructions: their History and Functions (Ithaca, N. Y., 1887), 3 ff.; German translation, 1 ff. 2 Cf. p. xix, n. 9. ' Cf. Haase, Ad Reisigium n. 504: "In der Volkssprache, die nichts leichter ver- nachlaessigt, als die Abhangigkeitsverhaltnisse der Saetze" (cited more fully, p. xviii, n, 5). * For the history of this theory, and its connection with the one just mentioned, see Hale, A Century of Metaphysical Syntax 199, and "The Heritage of Unreason in Syntactical Method," Proceedings of the Classical Association {of England and Wales) V (1907), S3 ff. For the refutation of the theory that the subjunctive is the mood of thought, cf. Hale, Cum-Constructions, 5 ff.; German translation, 3 ff. xvii xviii The Indicative Indirect Question in Latin indicative indirect question has been held to express something as cer- tain, or as a fact, and the subjunctive to express a mere thought.^ Closely connected with these theories about the meanings of the moods, is the notion that Latin syntax, and particularly that of the Ciceronian age, is something fixed and stereotyped.^ If the subjunctive was used in a given construction, some grammarians think, it had to be used in aJl examples of this construction. That Latin, and particularly the Latin of Cicero's age, might admit of either the subjunctive or the indicative in an identical use, is a thought that is repugnant to these scholars. Especially, that the same writer would employ now the one mood and now the other without any distinction in meaning, is an idea to which some minds, whether consciously or unconsciously, are closed.' ' Cf . Weissenbom, Syntax Latina 327 (quoted by Becker, op. cit. infra, 116): "So werden auch im Lateinischen . . . nicht selten solche Nebensatze als gewisse im Indikativ dargestellt." Cf. also Lindemann (cited by Holtze, op. cit. infra, 112): "Tenendum tamen, indicativum in his formiilis ubique habere aliquam confidentiae vel certae persuasionis notionem, ut non idem plane significent scio quid velis at scio quid vis"; and Paetzolt, De Latini Pronominis Relativi Syntaxi Prisca (Breslau, 1873), 26: "Ut quae rem complectantur factam vel vere extantem, in iis indicativus adhibeatur, coniimctivus ponatur in iis quae cogitati aliquid vel ficti contineant." Reisig, Schol. Lat. §329 (quoted with approval by Becker, 304) shows a combination of the theory that the subjunctive is the mood of thought with the theory that it is the mood of dependence: "Was aber hier [i.e., in interrogationibus obliquis] als ein subjektiver Gedanke dargestellt wird, kann auch in der Form der Objektivitat gegeben werden, wo die Sache nicht mehr in der Abhangigkeit von dem Denken steht, sondem fiir sich der grammatischen Form nach hingestellt ist als etwas Objektives. Hier ist also selbst die grammatische Form eine unabhangige, und der Indikativ wird dann ange- wendet, wofem nicht zu dem Inhalt des Gedankens der Sinn der Moglichkeit kommen soil." A similar combination appears in Haase's remark, Ad Reisigium n. 504 (quoted with approval by Becker, 118): "Viehnehr haben ihn [i.e., diesen Sprachgebrauch] die Komiker wohl schon vorgefunden in der Volkssprache, die nichts leichter vemach- lassigt, als die Abhangigkeitverhaltnisse der Satze; die lebendige Anschauung des Faktischen und die Neigung der Phantasie sich auch das nicht Faktische als solches vorzusteUen, veranlassen den Indikativ." * Cf. Becker, op. cit. infra, 115: "Nam in his ipsis legibus variandi studio multa commutata sunt, donee ad absolutam illam et certam loquendi rationem perventum est quam Ciceronis aetate valere constat" (the italics are mine). Cf. also Haase, Ad Reisigium n. 504 (quoted by Becker, 119): "In sorgfaltiger Redegatting, wo die grammatischen Verbindungen so pedantisch wahrgenommen werden, wie es in der besten Zeit der Romer immer geschah." ' Cf., for example, Fuhrmann, "Der Indikativ in den sogenannten indirekten Fragesaetzen bei Plautus," N. J.B. fiir Philologie u. Padagogik, CV (1872), 809-831, Introduction xix A priori theories concerning the meanings of the moods and the belief in a stereot3T)ed Latin syntax have combined to create a prejudiced attitude toward the indicative indirect question. Forced distinctions have been drawn between indirect questions with the subjunctive and apparent examples with the indicative;^ apparent examples of indicative mdirect questions have been interpreted as direct questions or as relative clauses;^ the construction, being regarded as something foreign to Latin, has been said to be due to Greek influence;" and, finally, recourse has been had to emendation.^! Holtze's work, Syntaxis Pnscorum Scriptorum Latinorum usque ad Terentium (Leipzig, 1861-62) may be cited in illustration of the preju- diced attitude toward the indicative indirect question. Holtze, Vol. IT, p. 110, quotes with approval the remark of Haase, Ad Reisigium, pp. 597 £f., n. 504, to the effect that the indicative indirect question originated in the popular speech, "in quo enuntiationum unam ab altera pendere facile negligatur et quae e sola cogitatione suspensa sunt, saepissime especially 810. Langen, "Commentationes Cornificianae," Philologus, XXXVII (1877), 405, emends the few cases of the indicative indirect question in the Ad Heren- nium, for the reason that the subjunctive indirect question is very frequent in that work. Similarly Draeger, Historische Syntax der lateinischen Sprache (Leipzig, 1881), 467, would emend the comparatively few cases of the indicative indirect question in the prose works of the younger Seneca. Kroll, "Der lateinische Relativsatz," Glotta, m (1910-12), 6, thinks that indicative indirect questions in Cicero are rightly emended, "deim Cicero setzt aus iibergrosser Gewissenhaftigkeit den Konjunktiv auch Ep. Ill, 10, 11." ' So Ramshorn, Lateinische Grammatik (Leipzig, 1830), II. 712, explains the use of the indicative in Cic. Verr. Act. II, III 26, 64, lam omnes intelligunt, cur universa provincia defensorem suae salutis eum quaesivit, on the ground that the indirect question expresses something as a fact: "wirklich gesucht hat." 'Hand, Tursellinus (Leipzig, 1829-45), I, 359 denies the use of the indicative in indirect questions and expresses the opinion that the apparent examples of this con- struction are to be explained as really being direct questions. For the interpretation of an apparent instance of an indirect question, as a relative clause, see the next para- graph. '» So by Reisig, Schol. Lat. §329 (quoted by Becker, p. 116). Cf. also Kiihnast, Die Hauptpunkte der limanischen Syntax (Berlin, 1872), 234: "Ein entschiedener Gracismus, und ohne ParaUele vor Livius ausser in der Komodie . . . ist der Indikativ in der indirekten Frage." " Cf . p. xviii, n. 7. The emendation of indicative indirect questions began at least as early as the time of Lambinus. Cf. Schmalz, Lateinische Syntax (Munich, 1910), 516. XX The Indicative Indirect Question in Latin tamquam facta animo praeponantur, " and says that some examples can be explained as direct questions, some as exclamations, and some as relative clauses. He does not, to be sure, state that all apparent examples of the indicative indirect question are to be explained away. However, he gives a forced interpretation when he remarks, p. 110: " Sic relativum esse videtur Cato R. R. c. 6: Vineam quo in agro conseri oportet, sic observato, h. e. in eo agro, in quo (ut iam scies), vinea conserenda est, haec observato. " The most comprehensive study of our subject which has yet appeared, Becker's De Syntaxi Interrogaiionum Obliquarum apud Priscos Scriptores Latinos, in Studemund, Studien I, (Berlin, 1873), pp. 113-314, likewise shows the influence of preconceived ideas. The purpose which Becker set before himself was to discover the laws^^" which in early Latin governed the use of moods in indirect questions. This purpose was, of course, perfectly legitimate. Only, an investigator should determine laws by the inductive method, after an extensive and unprejudiced study of phenom- ena. This, it seems to me, Becker did not do. Setting out with the purpose of discovering laws, Becker makes generalizations which the facts do not warrant. See below, pp. xxi f . Becker's conclusion concerning the use of the indicative mood in indirect questions in early Latin is as foUows (p. 119) : " Indicativus enim ubi in interrogationibus obliquis occurrit, profectus est aut ex laxo enuntiati primarii et secundarii conexu, aut ex inclinatione quadam pronominis interrogativi ad verbum enuntiati primarii aut denique ex ipsa enuntiati secundarii natura, ubi res de qua agitur ita omni dubita- tione vacat, ut contra rationem grammaticam propter banc praedicati condicionem indicativus positus sit." His indicative examples are classed in two main groups, the first group embod5dng the first of the three princi- ples mentioned, and the second group the second and third principles. ^ This is apparent throughout Becker's study. Cf., e.g., p. 116: "Certas enim leges in his modis adhibendis omnino non secutos esse priscos scriptores fere omnes sibi persuaserunt grammatici, qui hucusque hanc quaestionem tetigerunt: cui rei id praecipuo argumento est, quod non singula interrogationum genera distinxerunt, quae certis legibus aut indicativum aut coniunctiviun praeberent, sed in singulis exemplis indicativum excusabant"; and p. 119: "Quamquam enim apparebit in universum scriptores priscos easdem leges in adhibendis modis secutos esse atque posteriores, et quamvis saepe difficile sit legum, quas in ponendo aut indicative aut coniunctivo secuti sint scriptores prisci, certos constituere fines, tamen aliquot sunt interrogationum species in quibus certas leges in adhibendis his modis secuti esse videntur." Introduction xxi In the first group (pp. 120 S.), the indicative mood is explained on the ground that the connection between the clauses in the sentence is loose. Becker says, p. 165: "Docuimus autem indicativum inde explicandum esse, quod duo enuntiata tam laxe cohaererent, ut utrumque fere per se stare videretur. " Very many of the passages cited can, I agree with Becker, be well explained in such a way. Compare, e.g.. Plant. Bacch. 600 : Die mihi quis tu es (cited by Becker, p. 126). I should omit " fere " from the above quotation, and grant that the questions in Bacch. 600 and in many of Becker's examples may perfectly well be direct. How- ever, I cannot follow Becker when he attempts to set up rigid laws for the use of the indicative and of the subjunctive. He is governed largely by the assumption that where the indicative is used, the subjunctive could not be, and vice versa. Thus, having found a large number of examples of the indicative in questions accompanied by verbs in the imperative mood, he concludes that the use of the indicative in such questions is a law of early Latin, and emends the comparatively few examples that show the subjunctive (cf. his pp. 147 fE., and esp. 157fE.). Again, having set up the rule (p. 165), that when the connection between a question and its introductory verb is close, the question must be in the subjunctive mood, Becker emends the examples which contra- dict the rule. Thus he emends the indicative in questions that show "artificial prolepsis"^^ (pp. 168 ff.). He emends, likewise, the instances of the use of the indicative in clauses "in quibus is qui loquitur non alterum quendam interrogat sed aut ex ahquo, quern tamen non adloqui- tur, postea se aUquid quaesiturum esse praedicat aut hoc tantum indicat se id agere, quomodo de aliqua re certior fieri possit " (cf . his pp. 188 and 195). Similarly, he emends (p. 200) those examples which contradict his rule (p. 198) that " coniunctivus in eis interrogationibus flagitatur, ubi alium quendam id quod enuntiato secundario exprimitur rogitantem vel sciscitantem induco." For other passages which Becker emends because they do not comply with his laws, cf . his pp. 206, 219, 228, 240, 244-5, 262." •^ For "artificial prolepsis," see this study, p. 87, n. 1. "Becker has been criticized by various scholars for emending too freely: by Ussing (ed. Plautus, Hafnia, 1875-86), on Amph., Prologue 17; by Lindskog, Q^laes- Uones de Parataxi et Hypotaxi apiid Priscos Latinos (Lund, 1896), 92; by Antoine, "De la parataxe et de I'hypotaxe dans la langue latine," Reme des Hides anciennes, I (1899), 255 fi.; by Kroll, Der latemische RelaUvsatz, 5, n. 1. xxu The Indicative Indirect Question in Latin In Becker's second gro\ip of examples the indicative mood is explained on the twofold ground that the clauses are not mterrogative but relative, and that they express certainty or objective reahty (pp. 303 ff.)- As regards the explanation of apparent examples of the indicative indirect question on the ground that they are relative clauses, I agree with Becker that this is in very many cases possible. An instance is Plant. Amph. 460 : Ibo ad portum atque haec uti sunt facta ero dicam meo (cf . Becker, p. 306). Sometimes, however, Becker's use of this explanation results in forced interpretations. This is true, for example, in the case of Plaut. Rud. 965 : Et qui invenit hominem novi et dominus qui nunc est scio; and the similar examples in Rud. 958 and 1297 (Becker, p. 310). It is true also of Ter. Hec. 472 (p. 309) : Idque si nunc memorare hie velim, Quam fideli animo et benigno in illam et dementi fui, Vere possvun. The theory that the indicative is the mood of objectivity or certainty, and the subjunctive the mood of subjectivity or doubt, passed down to Becker from the metaphysical speculations of earlier scholars. Cf. p. xvii n . 4. This theory has been refuted, at least for the subjunctive mood, by Hale, Cum-Constructions, pp. 5 ff. (German translation, 3 ff.). The distinction which Becker draws, p. 304, is quite fanciful: "Quod dis- crimen intercedat inter enuntiata ilia, quae indicativum praebeant, et ea, quae coniunctivum, ex huiusmodi exemplis optime intellegitur, ut Amph. 1129, Simul hanc rern, ut factast, eloquar, et Trin. 236, Omnium primum Amoris artis eloquar, quemadmodum se expediant. In priore enim versu is qui loquitur indicativo ostendit se rem ita ut sit, expUca- turum esse, neque suam ipsius opinionem admiscentem neque rem cum aUis condicionibus mente conectentem; in posteriore vero enuntiato conixmctivo exprimitur rem non simpliciter narrari, sed ita exponi, ut eam esse is qui loquitur ex sua ipsius sententia sibi persuaserit : profitetur igitur, quemadmodum amoris artes se expediant, ita se elocuturum esse, ut ipse illas sese expedire compertum habeat, neque negat fieri posse ut alia quoque ratione illae sese expediant. " However, as Becker does not make much use of the theory of objectivity, but employs it only in con- nection with the explanation of clauses as relative, it is not worth while to dwell upon this point. Though Becker's aim is the establishment of rules for the use of the indicative and of the subjimctive in the examples — real or apparent — of the indirect question, yet he concedes to the early writers of Latin some degree of freedom. Thus he says, pp. 313 ff. : " In hac quaestione videre Introduction xxiii licet ... certum quidem sensum vel instinctum eos secutos esse in oratione concinnanda, at tamen ab experimento omnia nata esse neque deesse locos, in quibus ingeniosae licentiae nimis indulserint. " StUI, immediately after making this concession, Becker concludes his study with the following statement : " Id certe his plagulis effecisse mihi videor, ut in posterum ne ad taedium usque repeteretur decantata iUa modorum confusio, qua tum coniunctivus tum indicativus mirabili neglegentia effudisse dicebantur prisci scriptores. " It does not seem to me that Becker's conclusion is sound. It is only by the help of forced interpretations and of a considerable amount of emendation that he estabhshes his distinctions between the indicative and the subjunctive in indirect questions. A conclusion which involves such expedients is of questionable validity. Becker has exerted a powerful influence upon Latin grammars and syntactical studies, and upon editions of Latin authors. Draeger's treat- ment of the indirect question in early Latin is a summary of Becker's article.^^ Kuehner, likewise, closely follows Becker," and the Kuehner- Stegmann grammar shows his influence.^' Schmalz expresses the ideas of Becker and his predecessors, Lateinische Syntax, p. 516: "Im Alt- lateinischen hat sich der Indikativ in der indirekten Frage noch vielfach erhalten, d.h. die Stellung des Fragesatzes zum Hauptsatze ist eine ziem- lich selbstaendige . . ."; and again, " Selbstverstaendhch ist der Konjunktiv, wo der Inhalt des Gefragten noch zweifelhaf t und ungewiss ist, z. B. renuntiet mihi, velitne an non, oder wo bloss iiber eine Frage gesprochen wird, z. B. Cato Agr. I Praedium quod primum siet, si me rogabis, sic dicam, oder wo der Fragesatz infolge einer Prolepsis seines Subjekts — welche oft bei den Komikern vorkommt — die wesentliche und notwendige Erganzung des Hauptsatzes bildet, z. B. Plaut. Pers. 635 Patriam te rogo quae sit tua. " The explanation of apparent instances of indicative indirect questions as being really independent is given by Gildersleeve-Lodge^' and Lane.'' Lindsay, Syntax of Plautus (Oxford, " Historische Syntax (Leipzig, 1881), II, 460 ff., §§463 fE. " AusfuhrUche GrammaHk der lateinischen Sprache (Hanover, 1879), II, 989 £E., esp. 990 n. " AusfuhrUche Orammatih der lateinischen Sprache (Hanover, 1914), II, 2, 488 ff., §§227 ff. " Latin Grammar (New York, 1894), §467 n. " Latin Grammar (New York, 1898), §1787. xxiv The Indicative Indirect Question in Latin 1907), § 281, p. 66, where he refers to Becker's article, makes the state- ment that parataxis is in evidence "most of all in indirect questions, which in Plautus are as often direct (with indicative) as indirect (with subjunctive)." (The italics are mine.) The influence of the theories of Becker and his predecessors upon editions of Latin authors is apparent not only from explicit references to Becker like the one of Soimenschein on Rudens 385 (Oxford, 1891), but also from the common explanation that the indicative in indirect ques- tions is due to parataxis.^" This influence is observable, also, in the attempt to turn indirect questions into relative clauses by changing quid to qtiod.^ Becker concerned himself only with early Latin. However, he and his predecessors have strongly influenced the study of the Latin of later periods. Draeger, e.g., in his paragraphs upon the indicative indirect question in Ciceronian and early imperial Latin, employs the same methods as Becker.^^ The metaphysical theory that the subjunctive and indicative express, respectively, subjectivity and objectivity, is presumed in Postgate's note on Propertius 12,9 (ed. London, 1897) : "Some manuscripts have summittit. If right, the change of mood, summittit . . . wemaw/, is not due to any egsential difference of meaning, but is a reUc of a time when, as in Old Latin, the distinction between facts regarded as facts and as conceptions had not been evolved." The explanation that the indicative is due to parataxis is found in Appel, Beitrdge zur Erkldrung des Corippus (Munich, 1904), pp. 54 ff. : "Den auf die volkstiimliche Bevorzugung des parataktischen Satzverhaltnisses sich griindenden Indikativ in abhangigen Fragesatzen, der durch alia Zeitalter der romischen Volkssprache herrscht . . . bietet Corippus oft. " Bonnet, Le latin de Gregoirede Tours (Paris, 1890), p. 676, remarks : "II est probable que la veritable raison de I'indicatif c'est qu'on oublie que ces phrases sont des interrogations. On les confond avec les pro- positions relatives, conditionnelles, etc." ^ Cf. Morris' edition of the Captivi and Trinummus (Boston 1898), Introduction, §27: "These [i.e. indirect questions] are still largely in the paratactic stage." Cf. also Morris on Pseud. 262 (Boston, 1895); Sonnenschein on Most. 149 (Oxford, 1907); Ehner on Terence Phorm. 358 (Boston, 1895). 2' Cf. Tyrrell on Adel. 996 (ed. Terence; Oxford, 1902). » Op. cit., II, 473 ff., §§464 fi. Introduction xxv It is a significant fact, that the editors who nominally adopt the explanation of indicative questions as paratactic, almost without excep- tion show by their punctuation that they actually feel many of these questions to be hypotactic. Cf . the passages cited, p. xxiv, n. 20, from Morris' edition of the Pseudolus and Sonnenschein's of the Mostellaria. Indeed Becker himself punctuates most of the examples in his first group, not with interrogation-points, but with periods. See his pages 125 ff.^ Further, the acceptance of Becker's theories is often qualified by statements which reaUy contradict those theories. So Sonnenschein, on Plant. Most. 149, remarks that sometimes "indicative and subjunctive stand side by side without any clear logical distinction." Similarly Praun, Bemerkungen zur Syntax des Vitruvs (Bamburg, 1885), p. 71, after explaining the use of the indicative in indirect questions on the ground that " statt des Frageverhaeltnisses die relative Beziehung betont wird oder . . . der Gedanke als objektiv wahr hingestellt ist, " adds immedi- ately: "Wie geringfugig der Unterschied ist geht daraus h'ervor, dass nach dem gleichen iibergeordneten Verbum unter denselben Verhalt- nissen bald der Indikativ, bald der Konjunktiv folgt, ja dass sogar oft in einer Periode der Modus wechselt. " The clash between theory and feeUng is manifest also from the expression, "indicative questions ap- parently indirect" (cf. Lane, § 1787; Elmer, on Ter. Phorm. 358), and from the statement in Kiihner-Stegmann, II, § 227, 6a: "Die alte Sprache neigt hier (i.e., after qui, ut, quomodo, quamobrem, quam, etc.) zu relativer Auffassung und setzt demnach den Indikativ, wenn auch fast ueberall urspriingUch eine indirekte Frage zugrunde liegt und vielfach ' trotz des Indikativs eine relative Auffassung geradezu unmoglich ist. (The itaHcs in. both citations are mine.) There are, and there have been in the past, a considerable number of scholars who frankly accept the indicative indirect question in Latin as " To be sure, Becker is saved from actual self-contradiction by the fact that he regards the clauses of his first group, not as quite independent, but as "fere" indepen- dent (Becker p. 165; cf. my p. xxi). Similarly Elmer, on Ter. Phorm. 358, explains the indicative on the ground that the question is "not far removed from parataxis." Ehner, quite explicitly, and Becker, by implication, assume three stages: "Vide! Avaritia quid facit!"; "Vide avaritia quid facitl"; "Vide avaritia quid faciat!" How- ever, neither of these scholars undertakes to explain just ;svhat distinction in feeling there is between the second and the third stages, or — to put the matter a little differ- ently — at what degree of "removal from parataxis" the subjunctive mood displaces the indicative. - xxvi The Indicative Indirect Question in Latin an alternative construction to the subjunctive indirect question of fact. So Otto Schulz, Ausfuhrliche lateinische Grammatik (Halle, 1825) p. 480, says: "Es giebt allerdings Beispiele, wo auf ein Fragewort mit welchem eine indirekte Frage eingeleitet wird, der Indikativ folgt." Usually the acceptance of the indicative indirect question is a quaUfied acceptance. Thus J. N. Madvig, Lateinische Sprachlehre fur Schulen (Braunschweig, 1844), remarks, § 356, n. 3: "Bei den altesten Dichtern (Plautus und Terenz) steht bisweilen ein abhangiger Fragesatz im Indikativ . . . , bei den spatern (Horaz, Vergil) ist dieses selten, in Prosa ganz ungebrauchlich. "^^ 0. Wolff, De Enuntiatis Interrogativis apud Catullum, Tibullum, Propertium (Halle, 1883), p. 39, accepts the use of the indicative mood in indirect questions in poetry. He says: "Quam tenue sit ac paene nuUum inter coniunctivum et indicativum discrimen, luculentum exemplum praebet Prop. Ill 5, 27 ff. [cited, p. 17], ubi agnoscemus, quam hbere versatus sit in modorum commuta- tione. . . . Omnem operam quamquam dederam ut invenirem, quibus condicionibus indicativus, quibus coniunctivus positus videretur, tamen oleum et operam me perdidisse intellegere coactus sum, quod ad certam regulam utriusque modi usus redigi nuUo pacto potest. " Riemann-Goelzer, Grammaire comparie du grec et du latin (Paris, 1897), § 407, remarque I, acknowledges the use of the indicative mood in indirect questions " dans la langue vulgaire et dans la langue poetique, " but calls it an "incorrection. " Somewhat similar in tone is the comment of Allen and Greenpugh on Ovid, Met. X 637 (ed. Boston, 1890) : "The indicative in an indirect question is common in early Latin but is almost inexcusable in Ovid." Hallidie, Laing, and Sonnenschein acknowledge the use of the indicative indirect question in early Latin. Hallidie comments on Plaut. Capt. 207 (London, 1891): "The use of the indicative in dependent questions is not uncommon in Plautus. Setting aside those passages in which the indicative can be explained by taking the clause as an independent question or exclamation . . . there yet remain many instances in which the indicative is used although the clause is certainly dependent." Laing comments on Ter. Phorm. 358 "Becker, op. cit., 116, quotes with disapproval this part of Madvig's discussion. He remarks, 117 ff.:"Apparet . . . viros doctos de modis in interrogationibus obliquis adhibendis certas leges apud priscos scriptores nondum agnovisse." Introduction xxvii (Chicago, 1908) : " The indicative is frequently used in indirect questions in early Latin. " E. A. Sonnenschein, A New Latin Grammar (Oxford, 1912) remarks, p. 172, n. 2: "The use of the subjunctive in dependent questions as to a matter of fact . . . did not become a rule of Latin syntax till the time of Cicero, though it is often found in Old Latin. " Appel, op. cit., p. 55, n. 1, calls attention to the frequency of the indicative indirect question in late Latin. The Hale-Buck Latin Grammar (Boston, 1903) states, § 537 g: "The original indicative is still sometimes found in indirect questions ... in poetry (especially in early Latin), and in late colloquial prose." Examples are cited from Plant. Aul. 174 and Virg. Aen. VI 779. Marx, in his edition oi Ad Herennium (Leipzig, 1894), retains the indicative indirect questions "contra grammaticorum praecepta" (p. 176), and Schmalz, in spite of his acceptance of some of Becker's theories (cf . above, p. xxiii), commends Marx and expresses the opinion that "in all den Schriften auchderkIassischenZeit,welchederVolkssprachenahe stehen, der Indikativ der Ueberlieferung vielfach zu erhalten ist" (p. 516). Similarly Lejay, on Horace, Ep. I 7, 39 (Paris, 1912), conaments that the use of the indicative mood in an indirect question is "un trait de la langue f amiliere. " Cf. further, for statements, variously qualified, of the occurrence of the indicative mood in indirect questions in Latin, Riemann, La langue et la grammaire de Tite Live (Paris, 1885), p. 301, n. 3; Antoine, Syntaxe de la langue latine (Paris, 1885), § 243, 3; Allen and Greenough, Latin Grammar (Boston, 1899), § 334 d; Bennett, Syntax of Early Latin (Bos- ton, 1910), I pp. 120 ff.; Kroll, Der lateinische Relativsatz, p. 5. In a series of articles, the earliest of which appeared in 1904, Felix Gaffiot has tried to disprove the existence of the indicative indirect ques- tion in pre-Augustan Latin. The attempt seems to me unsuccessful. References to GafiSot are scattered throughout this study. See esp. pp. 81-84, 99 ff., 104 ff., and Appendices III and IV. Among scholars who acknowledge the existence of the indicative indirect question, there is a difference of opinion as regards its origin. Becker, p. 119, rejects the theory "of Greek influence (mentioned above, p. xix, n. 10) and agrees with Haase that the use of the indicative mood in indirect questions had its rise in the popular speech (cf. p. xviii, n. 5). That the construction is a colloquiaUsm is held also by Rebling, Versuch einer Charakteristik der rbmischen Umgangsspracke (Kiel, 1873), p. 7; xxviii The Indicative Indirect Question in Latin Riemann, La langue et la grammaire de Tite Live, p. 301, n. 3; Gerard, "Le latin vulgaire et le langage familierdans les satires de Perse," Musee beige, I (1897), p. 87, n. 7. The theory of Greek influence still has its adherents, though chiefly as an explanation of the occurrence of the construction in Augustan and later writers.^^ On the other hand, the use of the construction by these writers is sometimes considered an archaism.''* Some modern scholars, regarding language from the historical point of view, recognize that what really demands explanation is the use, not of the indicative mood, but of the subjunctive, in the indirect question of fact." Cf. Delbriick, Vergleichende Syntax (Strassburg, 1900), vol. Ill, p. 287. The problem of the origin of this use of the subjunctive mood lies outside the scope of the present study. II. Purpose and Method of this Study The question of the use in Latin of indirect interrogative clauses in the indicative mood is one which is important for text-criticism, as well as for the formation of opinion upon the rigidity or flexibility of syntactical usage.^^ It seems desirable, then, because there is no general agreement upon the subject and because current theories lead to emendation and to forced interpretations, that the evidence be examined anew. This paper is an attempt at an objective study of the problem. The method of procedure was as follows. First, most of the Latin works which were written before the end of the reign of Augustus were read, and the apparent examples of the construction were collected. Then the attempt was made to interpret every one of the examples other- wise than as an indirect question: as a direct question, as an exclama- tion, as a relative clause, as a condition, or in any other conceivable way. ^ Cf., e.g., Norden on Aeneid VI 615: "Der Indikativ im abhangigen Fragesatze nach der Praxis sowohl des Altlateinischen als des Griechisclien. Da er sich auch bei Properz, Ovid, und Spateren findet ... so muss fur Vergil das Griechische als massgebend betrachtet werden." " Cf. Kom-Ehwald on Ovid Met. X 637; Lejay on Horace Serm. II 4, 38. On the other hand, in his note on Horace Ep. I 7, 39, Lejay calls the use of the indicative in an indirect question "un trait de la langue familiSre." "The indicative was the original mood in this construction. Cf. Delbriick, of. cit., supra, 286. " For the belief in the existence of a rigid Latin syntax, particularly in the Cicer- onian age, cf. p. xviii, note 6. Introduction XXIX It was found that some examples lent themselves more or less readily to such interpretation, while others could not possibly be explained as any- thing else than indirect questions. The passages were accordingly separated into two divisions: indeter- minate examples, and apparently certain examples of the indirect ques- tion. The indeterminate examples were further classified into five sub- divisions. The first included those instances which might be considered direct questions; the second those which might be regarded as exclama- tions; the third those which might be considered relative clauses; the fourth those which might be considered conditions. The few miscel- laneous examples which remained constituted the fifth subdivision. Sometimes it was found that an example might be put into more than one of the first four subdivisions. Such an example was cited only once, in that subdivision into which it fell most naturally or with the least forcing. The examples in the first four subdivisions were again divided, according as it seemed (1) indifferent whether they were interpreted as indirect questions or were interpreted in some other way, or (2) more natural to understand them as indirect questions than to understand them in any other way. At first the classification was made on the basis of the writer's linguis- tic feeling. It was realized, however, that this feeling was a faUible guide, and so a search was made for objective standards. The direct question, the exclamation, the relative clause, and the condition were studied for the hght which they might shed, and criteria for distinguish- ing indirect questions from those constructions were formulated. Then the classification of examples was revised by means of these criteria. It was found impossible to escape altogether from subjective methods in separating those instances which it seemed natural to interpret otherwise than as indirect questions, from those for which such interpretation seemed more or less unnatural, but still conceivable. There is no abso- lute distinction between a natural and a somewhat unnatural interpreta- tion. On the other hand, the group composed of the examples which ,must be interpreted as indirect questions was determined wholly by ob- jective standards.^' " To be sure, the validity of these standards and of their application depends upon the writer's range of observation. In so far as this is true, the subjective element is omnipresent. XXX The Indicative Indirect Question in Latin To supplement the study of the indicative indirect question in the Augustan and earher ages, a similar, but less comprehensive, study was made of the construction in later Latin. The study of the earlier period constitutes Part I of this book. The supplementary study constitutes Part II. The Hsts of indeterminate examples are not meant to be exhaustive. I have cited all the instances which I have found, of clauses which are most naturally interpreted . as indirect questions. However, the hsts of clauses which may equally well be interpreted as indirect questions or as instances of other constructions are merely illustrative. I have aimed to give a complete list of the certain examples of the indicative indirect question in the authors and the collections that I have read in their entirety. (These authors and collections are designated in the bibliography by asterisks.) I can, however, scarcely hope that I have succeeded. With the exception of Plautus, I have read my authors through once only, and though I have checked my collection by means of collections in various grammars and in studies of the syntax of individual authors, and by cross-references in editions of various works,^" I have doubtless missed some examples. Further, in collecting the in- dicative indirect questions in many authors,^' the lack of complete critical editions is a serious handicap. In making citations, I have attempted to indicate aU manuscript variations which are ' relevant to my problem. In choosing among variant readings I have, in part, followed the editions cited in the bibli- ography. However, I have not hesitated to form independent judgments, especially when editors dififer among themselves, or when they seem to me to reject readings decidedly favored by the manuscripts, or not to give sufficient consideration to the lectio difficilior. In such cases I have often cited the readings of the editors in my critical notes (which are adapted from the editions cited in the bibUography). It is hardly possi- ble that my choice of readings wiQ always be correct, even when the choice is not rendered especially difficult by disagreement among authorities as to the relative value of manuscripts. However, as I strive to indicate all the relevant manuscript variations that are cited in the editions,'* ™ The books used for this purpose are included in the bibliography of this study, but are not designated in any particular way. *i Notably in Cicero. Cf . C. F. W. Mueller's remark in his edition (Leipzig, 1893-98), Part III, Vol. II, p. IV. s'' In a few instances I do not cite the variant readings, but refer instead to critical notes in the editions. Introduction XXXI it is to be hoped that any errors in judgment will not seriously impair the trustworthiness of this paper. The examples are arranged according to the meaning (1) of the introductory verb or verbal expression and (2) of the introductory pronoun or conjunction; thus: I Ask V Hear II Inform VI See II B Discuss VII Concern II C Depict VII B Care III Find Out VIII Wonder III B Observe IX It is Incredible III C Count X No Verb Expressed (Chapter III D Determine, Judge Headings, etc.) III E Consider, Reflect XI Reproach IV Know 1 Quid 17 Quatenus 2 Quis 18 Quando 3 Quoia 19 Quam 4 Quae, etc. 19 B Quamdiu 5 Qui (Masc. Sing. Substantive) 20 Ubi 6 Qua Causa, etc. 21 Unde 7 Quo Modo, etc. 22 Uter, etc. 8 Quahs, etc. 23 Ut 9 Quahter 24 -Ne 10 Quantus, etc. 25 Num 11 Quot 26 Utrum 12 Quotus, etc. 27 An 13 Cur 28 Si 14 Qui (Adverb) 29 Ne 15 Quo 30 Various Connectives 16 Qua 31 Cormectives Uncertain The division according to the meaning ; of the introductory verb is of necessity sometimes rather arbitrary. Thus, scire may mean either " to know" or "to find out" (cf.. for the latter meaning, Ter. Hec. 874: Ere licetne scire ex te hodie, quid sit quod feci iboni, . . . ?), and sometimes the meaning may be indeteiminate. However, the Index Locorum will facihtate the finding of the examples. PART I-A STUDY OF THE CONSTRUCTION TO THE END OF THE AUGUSTAN AGE- CHAPTER I An Attempt at Defining the Direct and the Indirect Question Before trying to distinguish between direct and indirect questions, it would be well if one could have a clear idea of what the term "question" and the terms "direct" and "indirect," as applied to questions, mean. In the first place, what is a question? The answer is not easy. No satisfactory definition of the word "ques- tion" has ever, so far as I can find, been worked out. Nor can I frame one. Nevertheless, a discussion is here in place, and may perhaps lead some one else to, make the attempt with success. I have searched a considerable number of grammatical works for a definition, but could find none. Writers both on comparative grammar and on the grammar of particular languages have refrained from defining the word. The definitions given in dictionaries are imperfect. For example, in the Funk and Wagnalls New Standard Dictionary of the English Language (New York and London, 1913) we find, 5. -u. "question," the following: "An interrogative sentence calling for an answer; a request for information; an inquiry." "Interrogative sentence" and "inquiry" are, of course, synonyms, rather than definitions, of the term question. The definition, "a request for information," is' defective, for it includes too much. "Please explain to me the meaning of this word" is a request for information, but is not a question. A definition of the term "question" should specify in what way the request for information is expressed. 1 2 The Indicative Indirect Question in Latin E. T. Owen, in Interrogative Thought and the Means of its Expres- sion (Madison, Wis., 1903) gives a preliminary definition which seems to me correct as far as it goes. He says, p. 408: "Answering now the possible query 'What interrogation is,' I feel it safe to say, in a general way, that it is one of several linguistic means of inducing another mind to give particular information." In the development of this thought, however, I do not wholly agree with him. In par- ticular, he over-emphasizes the importance of "specially question- asking" words. He thinks that such words are used not only in questions of the type, "Who killed Lincoln?" but also in so-called sentence-questions: e.g., "Is Brown honest?" "It appears," he says, pp. 463 ff., "that, just as in 'Who killed Lincoln?' the 'Who?' is regarded as the specially interrogative symbol, so also in 'Is Brown honest?' the 'Is' may be regarded as the specially question-asking word." It seems clear, on the contrary, that in the latter sentence interrogation is expressed by the word-order and the inflection of the voice, and not by "Is." I agree with Paul, Prinzipien der Sprachgeschichte (Halle, 1898), pp. 121 ff., that there are various means of expressing interrogation: (1) inflection of the voice; (2) a question-asking word; (3) word-order. These means may be used singly or together. I should, then, define a question as a sentence which, by one or more of these means, conveys a desire for information.^ This definition combines Owen's preliminary definition of the question and Paul's enumeration of the means of expressing interrogation. It should be noted that "question," in the definition, as in the passages cited from Paul, Owen, and the Funk and Wagnalls dictionary, means question in the narrow sense, i.e., the direct, and not the indirect, question. The definition given is not perfect, for it is formal. It does not explain what a question is, either psychologically or historically. Even as a formal definition, it has faults. In the first place, it is hardly legitimate to use the term "question-asking word" in the definition of a question. In the second place, since the term "sen- tence" is used, it ought perhaps to be explained what definition of the sentence is adopted. However, the definition given is adequate • The definition is meant to apply to the Indo-Euiopean languages in general, not to Latin in particular. An Attempt at Definition 3 for the present study. Its purpose is to help in the discrimination of the indirect, from the direct, question. What we have been trying to define, and what the passages cited from Owen and Paul are concerned with, is the question in the narrow sense, i.e., the direct, and not the indirect, question. Let us try now to understand what an indirect question is. Scholars sometimes speak as if the difference between the direct and the indirect question were, that the former is independent and the latter dependent. Cf., e.g., Kiihner-Stegmann, Vol. II, Part II, p. 487, §226, .1: "Die Fragen sind entweder unabhangig (direkt) oder von einem iibergeordneten Satze abhangig (indirekt)." However, a direct question may be in a sense dependent, as in "He asked, 'What have you there?' " It may be, too, that an indirect question is sometimes independent. Cf. German, "Ob das wohl wahr ist."' Perhaps it would be more exact to say that a direct question expresses a judgment,' and an indirect question a mere conception. But just what kind of conception do6s an indirect question express? And what do the direct and the indirect question have in common, that they should both be called in part by the same name, question? Here again we face difficulties. If we were to say that a direct question expresses an interrogative judgment and an indirect question an interrogative conception, we should next have to define the term interrogative. I have defined a direct question as a sentence which by certain definite means expresses a request for information. Might an indirect question be defined as a clause which suggests — but does not express — a desire for informa- tion by certain definite means (which would still need to be deter- mined)? In that case, the term interrogative would signify, "ex- pressing or suggesting a desire for information by certain definite means." But does an indirect question suggest a desire for information? Or is such suggestion, when there is such suggestion, made by the introductory verb? It may be thought that the indirect question * Some scholars would doubtless explain such clauses as dependent, with ellipsis of the introductory verb. ' Owen, op. cit., 409, calls the (direct) question a judgment. 4 The Indicative Indirect Question in Latin in itself suggests a desire for information, and that the introductory verb confirms or negatives this suggestion. Scire volo quid putes might illustrate the former effect of the introductory verb, and Scire^ nolo quid putes might illustrate the latter. Quid putes in itself is precisely the same in the two instances. However, I doubt whether it is accurate to speak of the indirect question as having any particu- lar force in itself. The force of an indirect question, and of many kinds of dependent clauses, is determined very largely by the context. It seems, accordingly, that an indirect question must be defined, if it is to be defined, in relation to a context. One might then say: "An indirect question is a clause which gives the content of a ques- tion but does not itself express a desire for information." This would distinguish between the direct question in "Rogavit, 'Quid habes?' " and the indirect question in "Rogayit quid haberem." But the definition breaks down when it is applied to a sentence like "Sciebam quid Marcus fecisset." It is not true, or not necessarily true, that "Quid Marcus fecisset" gives the content of a question. "Sciebam quid Marcus fecisset" does not mean, "I knew the answer to the question about what Mark had done." A person may know, and say that he knows, what another person has done, without ever having had a question about that other person's action arise in his mind. Many things come to our knowledge without our inquiring about them. One may say, "Sciebam quid Marcus fecisset" without thinking of any question (in the narrow sense, i.e., any direct ques- tion) whether present, past, or future, and whether one's own or another's, and indeed without experiencing any feeling of inquiry. In other words, many so-caUed indirect questions are strictly not questions at all. They neither express nor imply desire for infor- mation. More or less awareness of this fact is indicated by the distinction that is sometimes made between "proper" and "improper" {eigent- liche and uneigentliche) indirect questions. Indirect questions that occur after verbs which show that some one is inquiring* are regarded as "proper" indirect questions; all others as "improper" indirect questions. So the dependent clause in "I asked what you were doing" would be of the former kind; that in "I knew what you were * These indirect questions are indirect quotations. Cf . p. 89, n. 5, An Attempt at Definition 5 doing" would be of the latter kind. Cf. Kuhner-Stegmann, Vol. II, Part II, §226, 2. Cf. also the references to earlier scholars in Becker, p. 212. The term "indirect question," we have seen, is commonly applied to a group of constructions which differ so widely from one another that it is difficult, if not impossible, to frame a definition which shall apply to them all. These constructions belong together historically, and it is right that they should be called by the same name. But the task of defining the name involves difficulties which have not yet been overcome. Though I have been able to give only an imperfect definition of the direct question, and have been whoUy unsuccessful in my attempt at defining the indirect question, yet the present chapter has, I hope, made it sufficiently clear what the difference between a direct and an indirect question is. A direct question expresses a desire for informa- tion. An indirect question does not express this desire. It may give the content of some direct question, but it does not necessarily do so. CHAPTER II Indeterminatk Examples: Indirect or Direct Questions I. Questions whose Interpretation is Indifferent There are many indicative questions in Latin which might well be interpreted either as direct or as indirect.* Tlmt this should be true is in harmony with the generally accepted theory that the indirect originated from the direct question." Examples follow. The list is not meant to be in any sense exhaustive, but consists merely of illustrations. I have punctuated the questions as indirect, though it would be quite possible to punctuate them as direct. I. Ask 13. Cur. Varro, L. L. VIII 38, 70: Si analogia est, inquit, cur populus dicit dei penates, dei consentes. . . . ? Item quaerunt, si sit analogia, cur appellant omnes aedem deum consentium et non deorum consentium. Item cur dicatur mille denarium, non mille denari- orum. > This fact has been repeatedly observed. Cf., e.g., Lindskog, Quaestiones, 71. DelbiUck, Vergleichende Syntax, Vol. Ill, 276, implies that there are degrees of close- ness in the logical relationship between a question and the verb used with it. How- ever, the existence of various degrees of lo^cal subordination need not concern us here. What we are studying is ssTitactical, rather than logical, relations. Syntactically every question belongs to one of three classes: it is dependent, independent, or inde- tenninate. • Cf., e.g., KChner-Stegmann, Vol. 11, Part II, §227, 2. For what seems to me the probable way in which the indirect question sprang from the direct, see my "Theory of the Origin of Hypotaris," Indogermanische Forschungen, XXXV (1915), 242. Questions whose Interpretation is Indifferent 7 The Goet|t-Schoell edition, as its punctuation indicates, takes the question cuf ttppellant to be directly quoted. On that interpretation, the mood of dicatur, in the following sentence, would express pro- priety. For the subjunctive of obligation or propriety, cf. Hale- Buck, Latin Grammar, §§512 ff. 25. Num. C. Licin. Macer Cal- vus (Meyet p. 476): Rogo vos, indices, num, si iste disertus est, ideo me damnari oportet. II. Inforu /. Quid. "' Plant. Amph. 421: Signi die quid est. As. 358 : Quid nunc consiii captandum censes dice. Aul. 645 : Atque id quoque iam fiet, nisi f atere L. S. Quid fatear tibi? Euc. Quid abstulisti hinc. M. G. 441: Die mihi Quid hie tibi in Epheso est negoti. Poen. 1027: Narra quid est. 2. Quis. Plaut. Bacch. 553: Obsecro hercle, loquere quis is est. Merc. 620: Die quis emit. 14. Qui (Adverb) Plaut. Bacch. 1157: Nihili sum. Ni. Istuc iam pridem scio. Sed qui nihili's id memora.' 20. Ubi. Plaut. Merc. 606: Dice, obsecro. Si neque hie neque Accherunti sum, ubi sum. 22. Uter, etc. Plaut. Aul. 321: Sed uter uostrorum est celerior, memora mihi. ' qui Guietus: quid cod. Nihile sit cod. (nihiU sit B). 8 The Indicative Indirect Question in Latin 24. -Ne. Cic. Att. XV, 13, 6: Sed perscribe, quaeso, quae causa sit Myrtilo (poenas quidem ilium pependisse audivi), et satisne patet unde corruptus.* It may be that a new sentence, a direct question, begins with et. Cf. p. 11. 25. Num. Cic. De Orat. II, 65, 261: Die mihi, inquit, M. Pinari, num, si contra te dixero, mihi male dicturus es, ut ceteris fecisti. III. Find Out 4. Quae, etc. Plaut. Cure. 543: Th. Scire uolo quoi reddidisti. Ly. Lusco liberto tuo. 13. Cur. Pomponius Bono- niensis, Ribbeck, II, p. 240, 102: Volo scire ex te, cur urbanas res desubito deseris.^ IV. Know 24.— Ne. Virg. Aen. II, 739: Hie mihi neseio quod trepido male numen ami- cum Confusam eripuit mentem. Namque avia cursu Dum sequor et nota exeedo regione viarum, Heu! misero coniunx fatowe* erepta Creusa Substitit, erravitne via, seu lassa resedit Ineertum, nee post oeulis est reddita nostris. Most editions give the above punctuation and evidently regard both the -ne clauses and seu . . . resedit as indirect « *pateat Mue. and other edd.: patet codd. • deseres Par. P. 'fato mi Ribbeck. Questions Naturally Interpreted as Indirect 9 questions depending upon incertum. A few punctuate with question-marks after suhstitit and resedit. So the Fairclough- Brown edition (Boston, 1913), with the comment: "The disjointed utterances express realistically mental agitation. The -we is appended to fato, because the phrase, in which that noun is so important a word, belongs in common to the three verbs follow- ing. The seu, used instead of an interrogative, implies closer connection between the last two verbs than between substitit and erravit." VI. See 25. Num. Plaut. Most. 472: Circumspicedum, numquis est Sermonem nostrum qui aucupet. VIII. Wonder 1. Quid. Plaut. Rud. 614: Sed quid hie in Veneris fano meae viciniae Clamoris oritur, animus miratur meus. 30. Various Connectives. Ter. Phorm. 234: Quid mihi dicent aut quam causam reperient Demiror. II. Questions Which, if an Indicative Indirect Question Exists, are More Naturally Interpreted AS Indirect, than as Direct 1. Introductory. There are a considerable number of questions which may be direct but are more naturally understood as indirect. Just why they are more naturally so understood, one cannot always say. Indeed, there is no clear-cut division between these examples 10 The Indicative Indirect Question in Latin and those which have just been cited (cf. p. xxix). There are, how- ever, a number of circumstances to which one may point as indica- tions that questions are indirect. Direct questions are of two kinds: some are quotations, and some are not. An example of a direct question which is a quotation, is, " 'Where have you been?' he asked." It will simplify our task if we consider separately (1) those clauses which, if they are not indirect questions, will be most naturally interpreted as direct questions that are not quotations; and (2) those clauses which, if they are not indirect questions, will be most naturally interpreted as direct ques- tions that are quotations. 2. Indications that Questions are Indirect, rather than Direct and Not Quoted. In order to show that a question is indirect, rather than direct and not quoted, it is sufficient to show that it is dependent; for a direct question is never dependent, unless it is a quotation (cf. p. 3). This should be borne in mind as we proceed. (1) In the first place, a question is very probably dependent if it intervenes between another question and the answer to this other question. Cf. Plant. Men. 207: Scin quid nolo ego te accurare? Er. Scio: curabo quae uoles. It is conceivable that when a person has said, "Do you know? What do I want you to attend to?," the reply "I know" should be given. However, this reply is more appro- priate to the question, "Do you know what I want you to attend to?" (2) If a question seems to interrupt the introductory clause, it is probably dependent. It will be shown below (p. 90) that a question which interrupts the introductory clause is dependent, but that it is not always possible to determine whether a question does interrupt the introductory clause. The examples in which it seems probable, but is not certain, that the question interrupts the intro- ductory clause, are cited here. An instance is Plant. Pers. 640: Sed tamen, virgo, quae patriast tua, age mi actutum expedi. Here it seems probable that sed tamen goes with age. . . . exfedi, and that the 5Mae-clause is, therefore, dependent. Still it is possible that sed tamen belongs to the jwae-clause and that this is, accordingly, a direct question. (3) The grammatical or logical relations in which the introductory verb stands to other verbs may favor the interpretation of a question Questions Naturally Interpreted as Indirect 11 as dependent. So in Cic. Rep. 1, 19, 31, Turn Tubero: Non dissentio a te, Laeli, sed quaero quae tu esse maiora intellegis, the compound structure of the introductory sentence favors this interpretation. In Plaut. Pseud. 18, the logical relation oiface cerium and iuvabo has the same efiPect: Face me certum quid tibist; Iuvabo aut reaut opera aut consilio bono. (4) The circumstance that the introductory verb is modified by an adverb or a phrase, may make it more natural to understand the question as dependent than as independent. Cf. Plaut. Bacch. 558: Mn. Nequam homost, verum hercle amicus est tibi. Pi. Tanto magis Die quis est. (5) If the verb that accompanies an indicative question intro- duces another question which exhibits mood-shift and which is, therefore, clearly indirect, it is natural to interpret the indicative question too as indirect. Cf. Ter. Phormio 380: Quem amicum tuum ais fuisse istum explana mihi, £t qui cognatum me sibi esse diceret. This is not a sure criterion, for a speaker may conceivably change from a direct to an indirect question. So in the Phormio passage cited, it is possible — though, as it seems to me, not probable — that Quem . . .ais. . .is a direct question, followed by a shift to an indirect question: explana (understood) qui. . .diceret. Of course, an indirect question is often followed by a direct question that has no verb accompanying it. Accordingly, in Cic. Att. XV, 13, 6 (cited p. 8) the indicative question may quite well be interpreted as direct. (6) With verbs or phrases that do not express a desire of the speaker for information, an indirect question yields smoother Latin than a direct one. Die may equally well accompany a direct or an indirect question. On the other hand, dicam or the like may con- ceivably be used with a direct question, but yields smoother Latin when used with an indirect question. An example is Ovid Rem. Am. 683: Sed quid praecipue nostris conatibus obstat, Eloquar exemplo quemque docente suo. In general, it is more natural to interpret a question used in such an example as indirect than as direct; that is, it is more natural to punctuate. Quid . . . obstat, eloquar, than Quid . . . obstat? Eloquar. 12 The Indicative Indirect Question in Latin 3. Examples of Questions More Naturally'' Interpreted as Indirect, than as Direct and Not Quoted I. Ask 1. Quid. Plaut. M. G. 809: Meminero: sed quid meminisse id refert ' ego te tamen. 4. Quae, etc. Cic, Rep. I 19, 31: Turn Tubero: Non dissentio a te, Laeli, sed quaero quae tu esse maiora intellegis.' 24. -Ne. Cic. Flac. 32, 80: lUud quaero, suntne^" ista praedia censui censendo, habeant ius civile, sint necne sint mancipi, subsignari apud aerarium aut apud censorem possint. It would be harsh to interpret this passage as showing a shift from a direct to an indirect question. Cf. p. 11, criterion 5. II. Inform 1. Quid. Plaut. Aul. 777: Sat habeo. Age nunc loquere quid vis. Capt. 964: Tandem istaec aufer; die quid fers, ut feras hinc quod petis. Epid. 274: Quin tu eloquere, quid faciemus?" ' The examples vary greatly in the degree in which it is more natural to interpret them as indirect questions. Some of the examples may almost as well be direct, as indirect, questions. Others are almost certainly indirect questions. In many of them, editors have "emended" the indicative to the subjunctive, because they felt that the questions were indirect. In most cases my feeling agrees with the editors'. I cite these passages here, because I do not wish to include in Chapter VIII any passages which are not quite certainly indirect questions. ' Lindsay, following Brix, inserts rogo. A is illegible. • intdleg*gS cod: intellegas edd. '" suntne P: sintne Camerarius, edd. "This is a question of deliberation. Cf. M. G. 1183 (cited below); Stich. 706 (p. 22); Lucil. 375 (p. 27); Lucan II, 682 (p. 113). (In two of the examples the present tense is used.) For the use of the indicative mood in direct questions of Questions Naturally Interpreted as Indirect 13 Men. 763 : Sed haec res mihi in pectore et corde cura est, Quidnam hoc sit negoti quod sic filia Repente expetit me, ut ad sese irem. Nee quid id sit mihi certius facit, quid Velit, quid me accersit}'^ M. G. 1183: Quid? Ubi ero exornatus-quin tu dicis quid fac- turus sum?" Poen. 1087: Mi. Festivum facinus venit mihi in mentem modo. Ha. Quid id est? Mi: Tua opust opera. Ha. Die mihi quid luhet. Profecto uteris, ut uoles, operam meam. Pseud. 18: Face me certum quid tibist; luvabo aut re aut opera aut consilio bono." Rud. 946 : Quin loquere quid uis. Rud. 1102: Eloquere quid id est? Quid negoti est, modo dice. Da. Gripe, aduorte animum. Tu paucis ex- pedi quid postulas. Tr. Dixi equidem, sed si parum intellexti, dicam denuo. deliberation, cf. Plaut., Capt. 535, Quid loquar? Qvdd fabulabor? Quid negabo aut quid fatebor? Cf. also J. P. Deane, "Deliberative Questions, Indicative and Subjunctive, in Terence," Proceedings of the American Philological Association XXI (1890), pp. XXXIII fE., and Hale-Buck §571. Apart from the present example and the four parallels cited, all of our instances of the indicative indirect question are clauses oifact. " quod veUt, quod. Fowler, accersat, Lambinus. This is the punctuation of Leo. A direct question, "Quid me accersit?" as Goetz-Schoell and Lindsay have it, seems to me possible, but very awkward. •'i«w A: JM» P, Leo, Goetz-Schoell, Lindsay. Cf. the comment on Epid. 274 above. "fo'WwVFZ. 14 The Accius, Ribbeck p. 201, 499: Ter. Phor. 1048: Ovid, Rem. Am. 683: 2. Quis. Plaut. Bacch. 558 3. Quoia. Plaut. Merc. 529: Merc. 722: 4. Quae, etc. Plaut. Pers. 640: Indicative Indirect Question in Latin I, Praesto etiam adsum. Exprome quid fers:" nam te longo itere cerno hue vadere. Tu tuum nomeB die mihi quid est." Sed quid praecipue nostris conatibus obstat, Eloquar exemplo quemque docente suo." : Mn. Nequam homost, verum hercle amicus est tibi. Pi. Tanto magis Die quis est. Ly. Nunc, mulier, ne tu frustra sis, mea non es, ne arbitrere. Pa. Die igitur, quaeso, quoia sum. Ly. Quoia ea sit rogitas? Do. Resciscam tamen. Ly. Vin dicam quoiast? Ilia — ilia edepol — vae mihi! Nescio quid dicam.'' Do. At ego patriam te rogo quae sit tua. Vi. Quae mihi sit nisi haec ubi nunc sum? Do. At ego illam quaero quae fuit. Vi. Omne ego pro nilo esse duco quod fuit, quando fuit. ^feras Buecheler. " quid est A quod est BCDP. This is apparently an indirect question with pro- lepsis. For prolepsis, cf. Lindskog, Quaesliones, pp. 75 £E. " Thus R. Edd. ^ve various readings. It would be harsh to interpret this quid- clause as a direct question. However, it would not be absolutely impossible; for a person may ask a question whose answer he knows, merely to bring a subject to the attention of his audience. »8 Becker (p. 285) would explain away the indicative indirect question by distrib- uting: "L. Vin dicam? D. Quoiast?" This pimctuation, like Goetz-Schoell's "Vin dicam? Quoiast?" — for which cf. the comment on Ovid, Rem. Am. 683, cited above — seems less natural than that of Leo and of Lindsay, which I have adopted. Questions Naturally Interpreted as Indirect 15 To. Ita me di bene ament, sapienter; atque equidem miseret tamen. Sed tamen, virgo, quae patriast tua age mihi actutum expedi." Ter. Phorm. 380: Adulescens, primum abs te hoc bona venia peto, Si tibi placere potis est, mi ut respondeas: Quern amicum tuom ais fuisse istum explana mihi, Et qui cognatum me sibi esse diceret.^" Cic. Att. II 10: Nunc fac ut sciam, quo die te visuri sumus.'* 7. Quo Modo, etc. Cic. Att. VIII 2, 2: Nihil arbitror fore, quod reprehendas. Si qua erunt, doce me, quo modo iikyAJ/iv effugere possum.'^ 14. Qui (Adverb) Plaut. Pseud. 866: Co. Habe modo bonum animum. Ba. Quaeso qui possum doce^ Bonum animum habere qui te ad me adducam 20. Ubi. domum. Plaut. Bacch. 203: Die ubi ea nunc est, obsecro. Merc. 602: Prius quam recipias anhelitum, Uno verbo eloquere ubi ego sum, hicine an apud mortuos. Merc. 901 : Die igitur, ubi ilia est. Pseud. 599 : Nimis velim Certum qui d mihi faciat, Ballio leno ubi^ hie habitat. " Becker, p. 138, cites this passage among the examples in which he accounts' for the indicative mood as due to parataxis. It seems to me most natural to regard the jMae-clause as an indirect question. Cf. p. 10, criterion 2. '" It is most natural to consider the indicative question as indirect, and parallel to the subjunctive one, qui . . . diceret. See p. 11, criterion 5. ^ simus Miiller and other edd.: sumus M. As Skutsch has pointed out, Glolta III (1912) 366, the indicative yields a better rhythm than the subjunctive. But cf. p. 106, n. 1. ^ possum M, passim Miiller and other edd. ^ possim P. doce om. A. " Becker, p. 311, interprets the !spiam consequitur prope nos.^' Vide num eius color pudoris signum usquam indicat.^^. *' estne codd., Lindsay: sitne Pylades, Leo, Goetz-Schoell. GaflSot's interpre- tation (Rev. de phil., XXVIII, p. 49) that the question is direct seems to me forced. ^'This is the punctuation of Leo and Lindsay. Goetz-Schoell, and likewise Gaffiot, Rev. de phil., XXVIII, p. 51, regard the question as direct. Gaffiot concedes that on this interpretation estne ibi is absurd, because the person addressed cannot have gone to the temple. However, he argues, "c' est pr6cis6ment dans cette absurdity que r&ide tout le sel . . ;" for the person at whom the joke is aimed is a soldier, of whom Plautus, as is his practice, makes ''un balourd inintelligent," incapable even of seeing that he is being openly ridiculed. Gaffiot's interpretation seems to me very improbable. "" This is the punctuation of Lindsay, Goetz-Schoell, and Leo. " This is the punctuation of Lindsay, Goetz-Schoell, and Leo. ''i This is the punctuation of Tyrrell and of Dziatzko (Leipzig, 1884). Fair- clough has ?. 24 The Indicative Indirect Question in Latin VII. CONCEKN 1. Quid. Cic. Att. V 20, 7: At te Romae non fore! Sed est totum quid Kalendis Martiis futurum est.^' VIII. Wonder 7. Quo Modo, etc. Plaut. M. G. 377: Nimis mirumst facinus quo modo haec hinc hue transire potuit. M. G. 418: Sed facinus mirum est quo modo haec hinc hue transire potuit.^^ 14. Qui (Adverb). Plaut. Epid. 414: Mirum hoc qui potuit fieri.*^ 15. Quo. Plaut. Stich. 541: Miror quo evasurust apologus.^° 4. Indications that Questions are Indirect, rather than Directly Quoted. (1) When one and the same verb introduces both a question which may conceivably be either directly or indirectly quoted, and also a question which must be indirectly quoted, it is most natural to consider the former question as of the same kind as the latter. Thus, in Cic. Inv. I 28, 43 (cited p. 27) "In hac eae res quaeruntur," is followed, first by subjunctive indirect questions, and then by "postea homines id sua auctoritate comprobare an offendere in lis consuerunt." The natural interpretation, it seems to me, is that this indicative question is, like the subjunctive questions preceding it, indirectly quoted. To be sure, a shift from indirect to direct quotation is conceivable.^' «' Thus Gaffiot, Pour le vrai latin (Paris, 1909) 71, vrfth the codd.: edd. "emend." It would perhaps be possible, though not natural, to interpret: "But the all-important question is: What is going to happen on the first of March?" " Becker's interpretation (p. 312) of these two examples as relative clauses seems to me even more forced than the interpretation that they are direct questions. " hoc qui ] F Pylades: hoc quodB: hoc quidem Z. •* Becker, p. 311, suggests a change to quorsum evadat. " For shifts from indirect to direct quotation, cf. R. Krumbiegel, De Varroniano Scribendi Genere Quaesiiones (Leipzig, 1892) S3 n. 3. Indications that Questions are Indirect 25 (2) If, from the context, it seems that the speaker's, or writer's, interest is merely in the content, and not at all in the form, of the question, the question is better regarded as indirectly, than as directly, quoted. E.g., in Plaut. Epid. 438, Cave praeterbitas ullas aedis, quin roges,/Senex hie ubi habitat Periphanes Platenius, the speaker would naturally be interested, not in the exact words in which the question is to be expressed, but merely in the content of the question. The question, is, therefore, more probably indirect than direct. (3\ After the verb mirari, it seems more natural to regard a question as indirectly, than as directly, quoted. Catullus 69, 10, Aut admirari desine cur fugiunt, may conceivably be understood to mean: "Or cease to wonder: 'Why do they run away?' " However, it is more naturally interpreted: "Or cease to wonder why they run away." (4) When a question shows a shift in person, it is natural to under- stand it as indirect. Instances of such questions are Lucil. 375 and Cic. Att. VIII 11, 5, cited below. It might be thought that a shift in person proves conclusively that a question is indirect. It is, however, conceivable that a clause like the ones cited may be a mixture of direct and indirect quotation. In English, for example, there occur expressions like "As for your question, what did Smith write to me?" In this expression the use of the person is appro- priate to indirect quotation; but the position of the verb, its form — "did write" for "wrote" — ,the more expressive delivery of the quota- tion, and the longer pause before it are appropriate to direct quota- tion.^' Similarly, it is conceivable that in "Quod quaeris, quid Caesar ad me scripsit" (Att. VIII 11, 5) "ad me" belongs to indirect quotation but the mood-use is due to the question's being felt, in some degree, as direct. However, this explanation seems somewhat forced. " An instance of such mixture of direct and indirect quotation is the followmg passage from Scott, The Heart of Mid-Lothian, Chap. 41 : "But when was she to see Butler? was a question she could not forbear asking herself." 26 The Indicative Indirect Question in Latin 5. Examples of Questions More Naturally Interpreted as Indirect, than as Directly Quoted. I. Ask 1. Quid. Ter. H. T. 1008: At si rogem iam quid est quod peccem aut quam ob rem hoc facias, nescias, In qua re nunc tam confidenter restas, stulta.'' Cic. Att. VIII 11, 5: Quod quaeris, quid Caesar ad me scripsit, quod saepe, gratissimum sibi esse quod quie- rim.^" Vitruv. II 6, 4: Relinquetur desideratio, quoniam item sunt in Etruria ex aqua calida crebri fontes, quid ita non etiam ibi nascitur pulvis. 4. Quae, etc. Cic. Tusc. I 13, 29: Si vero scrutari Vetera et ex iis ea quae scrip- tores Graeciae prodiderunt, eruere coner, ipsi illi maiorum gentium di qui habentur, hinc a nobis profecti in caelum reperientur. Quaere quorum demonstrantur sepulcra in Graecia, re- miniscere, quoniam es initiatus, quae tradantur mysteriis; tum denique quam hoc late pateat, intelleges.^i 6. Qua Causa, etc. Varro, Excerpta ex Augustini Libro qui est de Dialectica (Goetz-Schoell, Varro, De Lingua Latina, p. 341, 21): Scrutatur, ipsum vincire unde dictum sit: dicemus a vi. Vis quare sic appellatur, requiret. " Becker, p. 183 note, interprets "quid . . . facias" as a directly quoted question. '"scripsit M, Gaffiot {Pour le vrai latin 68): scripserit vulg.; scripserit, scripsit TyrreU-Purser. Cf . above, criterion 4. *' demonstrantur R B O: demonstrentur Augustine, de Cons. Evang. I 23, and (changed by the same hand to demonstrantur) G, and edd. Questions Naturally Interpreted as Indirect 27 7. Quo Mode, etc. Varro, Reliquorum de Grammatica Librorum Frag- menta68; Goetz- Schoell, p. 207; Keil G. L. VII 150, 10; Isid. I 27, 15: Lacrumae an lacrimae, maxumus an maxi- mus et si qua similia sunt quomodo scribi debent quaesitum est.^^ 20. Ubi. Plaut. Epid. 438: Cave praeterbitas ullas aedis, quin roges, Senex hie ubi habitat Periphanes Platenius.^' 24. -Ne. Lucil. 375 (Vel. Long. G. L. VII p. 62, 1 K.) : Atque accurrere scribes Dne an c non est quod quaeras eque labores.'^ 27. An. Cic. Inv. I 28, 43: In hac eae res quaeruntur . . . [subjunctive indirect questions]; postea homines id sua auctoritate comprobare an offendere in iis consuerunt; et cetera, quae factum aliquid similiter confestim aut ex intervallo solent consequi.^^ '2 debeant Carrio, edd. " Becker's interpretation (p. 311) that the clause is' relative, with locum under- stood, seems to me impossible. Rogo is never used, so far as I can find, with a direct object of the thing inquired about. Indeed Becker himself is not wholly satisfied with his interpretation; for he remarks. I.e., n. 3: "Praeterea nescio an . . . hahitet vel habeat scribendum sit." "scribas Dousa, Marx. This is apparently an indirect deliberative question. Cf. p. 12 n. 11. ^ 'Edatois consuerini 01 consueverint. Mtillern. cr.: "consuerunt HP; consue- veruntS; consueverint IP." Halm-Bai., n. cr.: "consuerint A T; consuerunt V ; con- sueverint V R E." 28 The Indicative Indirect Question in Latin VIII. Wonder 13. Cur. Lucr. IV 290: Quare etiam atque etiam minime mirarier est par, Illic quor reddunt speculorum ex aequore visum, Aeribus binis quoniam res confit utraque.^' CatuU. 69, 10: Quare aut crudelem nasorum interfice pestem, Aut admirari desine cur fugiunt.*' " lUic quor reddant edd. : Illis quae reddunt O Q. " cum Froehlich. frighmt O: fugiant A. It would seem to me even more harsh to interpret these car-clauses as relative clauses than to consider them directly quoted questions. CHAPTER III Indeterminate Examples: Indirect Questions or Exclamations' I. Clauses whose Interpretation is Indifferent Exclamations are often identical in form^ with questions, both with direct and with indirect' questions. The clause "Quid ille facit," for example, may be an instance of any one of these three construc- > The objection may be raised, that for the clauses studied in this chapter, the alternative interpretations are, not indirect question and exclamation, but dependent and independent exclamation. The point is not an important one, as far as this study is concerned. It would not affect our conclusions, if this subdivision were called, "Examples Which May be either Dependent or Independent Exclamations." I have chosen the phraseology, "Indirect Questions or Exclamations" for the reason that I do not think that a dependent clause can ever be properly called an exclamation, except when it is a quotation of some one's speech or thought. An example of an indirectly quoted exclamation is Virgil, Aeneid, I 454: "Namque sub ingenti lustrat dum singula templo Reginam opperiens, dum quaefortuna sit urbi Artificumque manus inter se operumque laborem ' Miratur, videt Hiacas ex ordine pugnas." The examples cited in this chapter are of a quite different character. Consider, e.g., Flaut. Stich. 410, Videte, quaeso, quid potest pecunia. If the gwid-clause is dependent, it is not, it seems to me, an exclamation. The sentence as a whole may be an exclama- tion, but not the dependent clause in itself. The force of the dependent clause in "Videte, quaeso, quid potest pecunia" is just the same as the force of the dependent clause in "Dicam quid potest pecunia" (an imaginary example). If the latter quid- clause is called an indirect question, the former quid-dz.use should be designated by the same name. ' In spoken language, to be sure, inflection of the voice, rapidity or slowness of speech, and the length of the pauses between clauses may differentiate the direct question, the indirect question, and the exclamation. ' In Latin, which makes no distinction in word-order between direct and indirect questions (cf. p. 91), a particular exclamation is often identical in form both with the direct and witii the indirect question. In some languages exclamations resemble at times direct questions and at times indirect ones. For the former kind of exclamation, cf. French, Avons-nom ril; German, Wiejreut sie sich darmfl. For the latter kind, cf. 29 30 The Indicative Indirect Question in Latin tions. Often the context will decide the interpretation. At times, however, it is impossible to distinguish between the indirect question and the exclamation. I have punctuated the following examples as indirect questions. However, they might equally well be punctuated as exclamations. VI. See 4. Quae, etc. Ter. Eun. 242: Viden me ex eodem ortum loco, Qui color, nitor, vestitus, quae habitudost cor- poris? Omnia habeo neque quicquam habeo.^ 7. Quo Modo, etc. Plaut. M. G. 201: Illuc sis vide, Quem ad modum astitit, severe fronte curans, cogitans. 19. Quam. Plaut. Capt. 557: Viden tu hunc, quam inimico uoltu intuitur? 23. Ut. Plaut. Men. 828: Viden tu illic oculos virere? Ut viridis exoritur colos Ex temporibus atque fronte, ut oculi scintillant, vide. 30. Various Connectives. Plaut. Most. 830: Specta quam arte dormiunt. Tk. Dormiunt? Tr. lUud quidem ut conivent uolui dicere. French, Comhien de pommes vous avez W; German, Wie schon diese Aussicht isti; English, How beautifid this view is! Just how the exclamation is related historically to the other constructions is not certain. Paul, Prinzipien, 123, thinks of exclamations as rhetorical direct questions. Brugmann, on the other hand, Kurze vergleichende GrammaHk, 942, shows how exclama- tions may develop from indirect questions. * Tyrrell begins a new sentence with qui, and places ! after corporis. Clauses Naturally Interpreted as Indirect Questions 31 II. Clauses Which, if an Indicative Indirect Question Exists, are More Naturally Interpreted as Indirect Questions than as Exclamations 1. Indications that Clauses are Indirect Questions rather than Exclamations. There are a number of examples of clauses which may be exclamations but are most naturally interpreted as indirect questions. The reasons why the interpretation of these clauses as indirect questions seems the most natural one are not always easy to formulate. There are, however, a number of circumstances which may be pointed to as favoring the interpretation that the clauses are indirect questions. These circumstances are in part identical with those which were found to favor the interpretation of questions as indirect rather than as direct (cf. pp. 10 ff.). Thus, (1) if the clause under consideration intervenes between a question and the answer to it, the clause is probably dependent. Cf. Plant., As. 884: Pa. Audin quid ait? Art. Audio. One might perhaps object that, in such an instance, the clause is a dependent exclamation, rather than a dependent question. But there is no such thing, except in quotations, as a dependent exclamation that is distinct from a dependent question (cf. p. 29, n. 1.). (2) When there is a close logical connection between the passage under debate and a preceding sentence, it sometimes yields the smoothest and most natural Latin, to regard the clause under con- sideration as dependent, and hence an indirect question. Cf. Plant., Stich. 310: Nimis haec res sine cura geritur: vide quam dudum hie asto et pulto. Here the words that follow the colon confirm the statement that precedes it. This confirmation is most smoothly and naturally made if vide is the principal verb and the quam-cla,use is dependent. (3) The fact that the introductory verb is modified by an adverb or a phrase may make it more natural to understand the clause under consideration as dependent, and an indirect question, than as independent, and an exclamation. Cf. Sil. Ital., XIII 446: Interea cerne, ut gressus inhumata citatos Pert umbra et properat tecum coniungere dicta.* ' This example, from Part II, is cited here, because there is no good illustration of the principle among the earlier examples. 32 The Indicative Indirect Question in Latin (4) The occurrence, next to the clause under consideration, of a clear example of an indirect question, to which it seems to be con- nected by a coordinating conjunction, favors the interpretation of the clause as an indirect question. Cf. Ter. Andr. 650: Ah, nescis quantis in malis uorser miser/Quantasque hie consiliis suis mihi con- flavit soUicitudines/Meus carnufex. (5) The meaning of some verbs makes it less natural for them to be followed by an exclamation than by an indirect question. The verb memini is an example. Cf. Cic. Lael. 25, 96: Atque, ut ad me redeam, meministis, Q. Maxumo . . . et L. Mancino consulibus, quam popularis lex de sacerdotiis C. Licini Crassi videbatur. These five criteria correspond, in a general way, to the first, third, fourth, fifth, and sixth criteria on pp. 10 fE. There are a few addi- tional circumstances which favor the interpretation of clauses as indirect questions rather than as exclamations. (6) With an introductory verb that is not in the imperative mood, it is generally more natural to utter an indirect question than an exclamation. Conversely, it is more natural to interpret an indeterminate clause used with such a verb, as an indirect question than as an exclamation. Cf. Plaut. Capt. 592: Heus, audin quid ait? (Cited more fu_lly, p. 35.) (7) In a passage that is calmly intellectual in tone, an exclamation would be out of place. An example is Virg. Georg. I 57: Nonne vides croceos ut Tmolus odores, /India mittit ebur, moUes sua tura Sabaei,/At Chalybes nudi ferrum. . . .? (8) The content of a clause is sometimes of a kind with which, in the context in which the clause occurs, exclamatory feeling does not harmonize. Cf. Plaut. Pseud. 330: Ba. Nolo victumas: agninis me extis placari nolo. Cali. Propera, quid stas? I accerse agnos. Audin quid ait luppiter? In this context, quid ait luppiter cannot well have any exclamatory tone. The clause is, in all probability, an indirect question. (9) Sometimes the structure of the clause under consideration opposes the interpretation of the clause as an exclamation. So in Plaut. As. 636, Videtin viginti minas quid pollent quidve possunt, the conjunction -ve is indicative of reflection, and, while it is con- ceivable that it should be used after an exclamation, its occurrence is Clauses Naturally Interpreted as Indirect Questions 33 more natural on the assumption that quid pollent is an indirect question. 2. Examples of Clauses More Naturally Interpreted as Indirect Questions than as Exclamations.^ II. Inform IQ. Quam. Ter. Hec. 472: Idque si nunc memorare hie velim, Quam fideli animo et benigno in illam et dementi fui, Vere possum, ni te ex ipsa haec magis veHm resciscere.' IIIB. Observe 1. Quid. Plaut. Men. 472: Observa quid dabo.^ 19. Quam. Plaut. Amph. 507: Observatote, quam blande muUeri palpabitur.' HIE. Consider, Reflect 10. Quantus, etc. Cato Orig. Fr. 95b (Ex Libro V) (Peter ed. 1914, p. 86): Cogitate quanta nos inter nos privatim cau- tiusfacimus. Nam unus quisque nostrum, siquis aduorsus rem suam quid fieri arbitratur, summa vi contra nititur, ne aduorsus eam fiat: quod ilU tamen perpessi.^" • The clauses cited in this section vary greatly as legards the degree in which it is more natural to interpret them as indirect questions. Some of them may almost as well be understood as exclaniations. Others are almost certainly indirect questions. ' It would be unnatural, but perhaps not wholly impossible, to regard the quam- clause as a directly quoted exclamation, in apposition with id: "Quam fideli animo . . . Ml" ' Thus BCD, Leo, Goetz-Schoell: observabo quid agat F Z. • Cum blande suppalpatur tmdieri schol. Virg. Aen. II 725: ut blande paipetur miMeri Donatus Adel. Prol. 2. i» Becker, p. 301, would ledAfaciamus. 34 The Indicative Indirect Question in Latin 19. Quam. Ter. H. T. 638: Quam bene vero abs te prospectumst quid uoluisti cogita: Nempe anui illi prodita abs te filiast planissume, Per te vel uti quaestum faceret vel uti veniret palam.^^ IV. Know 4. Quae, etc. Plaut. Trin. 283 : Novi ego hoc saeculum moribus quibus est.^^ 10. Quantus, etc. Plaut. M. G. 1075: Non edepol tu scis, mulier, Quantum ego honorem nunc illi habeo.'^ Ah nescis quantis in mails uorser miser Quantasque hie consiliis suis mihi conflavit soUicitudines Meus carnufex.i^ Ter. Andr. 650: 19. Quam. Cic. Lael. 25, 96: Atque, ut ad me redeam, meministis, Q. Maxumo . . . et L. Mancino consulibus, quam popularis lex de sacerdotiis C. Licini Crassi videbatur: cooptatio enim collegiorum ad populi beneficium transferebatur. . . . Tamen illius vendibilem orationem religio deorum' immor- talium nobis defendentibus facile vincebat.^^ " quid codd., Dziatzko: qiwd Bo the, Tyrrell. Becker, p. 302, is troubled by this example, and would emend it. ^siefP: et A. Studemund, Rheinisches Museum XXI (1866) 593: "Der Codex hat deutlich quibuset was wohl weniger auf 'quibus siet' als auf 'quibus est' hinzeigt." ^ Becker, p. 219, suggests a change to habeam. It would be extremely forced, indeed practically impossible, to interpret the qu^ntum-claMse as an exclamation. " Gaffiot, Rev. de phil., XXVIII p. 54, interprets the indicative clause as an exclamation. Becker's interpretation (p. 219) is similar. ''This is the punctuation of Halm-Baiter. Baiter-Kayser and Miiller place / after videbatur. The latter punctuation seems to me less natural. Clauses Naturally Interpreted as Indirect Questions 35 23. Ut. Plaut. Bacch Most. 202: Vide, quaeso, ne quis tractet illam indiligens: Scis tu ut confringi vas cito Samium solet." 149: Cor dolet quom scio ut nunc sum atque ut fui." 1. Quid. Plaut. As. 884: Capt. 592: Pers. 655: V. Hear Pa. Audin quid ait? Art. Audio.^' Heus, audin quid ait? quin fugis? lam illic hie nos insectabit lapidibus, nisi illunc iubes Comprehendi. Vi. lam hoc tibi dico: actutum ecastor mens pater, ubi me sciet Veniisse, ipse aderit et me abs te redimet. To. Quid nunc? Do. Quid est? To. Audin quid ait? Ba. Nolo victimas: agninis me extis placari uolo. Ca. Propera: quid stas? I accerse agnos. Audin quid ait luppiter? Audin tu, hie quid ait?i' " Becker, p. 254, suggests a change to "Scin tu? confringi" or the like. " It would be so forced as to be practically impossible, to interpret the «i-clauses in these two examples as exclamations: the former, "You know this: How quickly Samian ware is smashed!"; the latter, "It grieves me to think about it: what sort of person I am, and what sort I have been!" Becker's interpretation (p. 310) of the latter example as a relative clause is quite impossible: "Mente supple fere: Scio me taJem, ut. . . ." ** Becker, pp. 271 S., thinks that attdin has no real interrogative force and that the question is not really dependent. However (p. 285), he acknowledges that in Ter. Heaut. 731 a reply, Audivi — which corresponds to audio in this example — shows that the preceding atidisti — which corresponds to awlin in this example — ^is a real question, and that the following interrogative clause is dependent. Cf., against Becker, Morris, "On the Sentence Question in Plautus and Terence,'' A.J.P. X (1889) pp. 404 ff. Becker's interpretation is, to say the least, extremely unnatural. 19 Becker's explanation, pp. 271 ff., that these examples are exclamations — ■ "exclamationes interrogationum formam indutae" — is forced. In all these passages the natural interpretation is that the gwiif-clauses are indirect questions. For Pseud., 330, particularly, an exclamation would be out of harmony with the context. Cf. p. 32, criterion 8. Pseud. 330: Ter. Eun. 1037: 36 The Indicative Indirect Question in Latin 23. Ut. Plaut. As. 598: Audin hunc opera ut largus est nocturna? Men. 919: Audin tu ut deliramenta loquitur? Enn. ap. Non., p. 150, 6 (Annales VI; Vahlen 210) : Divi hoc audite parumper, Ut pro Romano populo prognariter armis Certando prudens animam de corpora mitto. VI. See 1. Quid. Plaut. As. 636: Videtin viginti minae quid pollent quidve pos- sunt?"" Pers. 292: Specta Quid dedero. Pa. Nihil, nam nihil habes. Stich. 410: Videte, quaeso, quid potest pecunia. Trin. 847: Viden egestas quid negoti dat homini misero mali? Ter. Eun. 265 : Viden otium et cibus quid f acit alienus?^^ Phor. 358: Quia egens relictast misera, ignoratur parens, Neclegitur ipsa: vide avaritia quid facit. Cic. Att. XIII 18, 1 : Vides propinquitas quid habet.^^ Caecil. Statins, Rib- beck II, p. 71, 216: Vide, Demea, hominis quid fert morum simili- tas.^' Prop. II 16, 29: Aspice quid donis Eriphyla invenit amaris, Arserit et quantis nupta Creusa malis.^ '" It is only by extremely forced interpietations that this example, and also some of the examples cited below, can be interpreted as anything else than indirect ques- tions. Becker, p. 272, considers that viden and the like are imperative in effect, and that the clauses following are exclamations. Becker's interpretation is opposed by Morris in A.J.P., X (1889), pp. 404 £E. '^facial A (according to Fabia, ed. Paris, 1895). ^ habet M, Gaffiot (Pour le vrai latin, 71) : habeat Lambinus, Muller, and other edd. ^fert edd.: feret codd. « Cf . p. 32, criterion 4. Clauses Naturally Interpreted as Indirect Questions 37 4. Quae, etc. Plaut. Most. 199: Postremo, si dictis nequis perduci ut vera haec credas Mea dicta, ex factis nosce rem. Vides quae sim et quae fui ante.^^ 7. Quo Modo, etc. Pers. 386: Tace, stulta, non tu nunc hominum mores vides, Quoiius modi hie cum mala fama facile nubitur? Dum dos sit nullum vitium vitio uortitur.'" 10. Quantus, etc. Ter. Phorm. 987: Os opprime inpurum: vide Quantum valet. 15. Quo. Ter. Eun. 238 : Quoniam miser quod habui perdidi, en Quo redactus sum. Omnes noti me atque amici deserunt." 19. Quam. Plaut. Stich. 310: Nimis haec res sine cura geritur: vide quam dudum hie asto et pulto. Ter. Hec. 223: At vide quam immerito aegritudo haec oritur mi abs te, Sostrata: Rus habitatum abii, concedens vobis et rei Servians, Sumptus vostros otiumque ut nostra res posset pati.^' Cic. Att. VIII 13, 2: Et vide quam con versa res est: ilium quo antea confidebant metuunt, hunc amant quern timebant.^' ^ Leo and Goetz-Schoell, following Seyfiert, place : after sim. Probably they think of quae fui ante as exclamatory, though they do not punctuate with an exclama- tion-point. Gaffiot, Rev. de phil., XXVIII, p. S3, interprets the clause as an exclama- tion. Becker, p. 297, takes it as an indirect question and proposes a change to fjierim. ™ cuiusmodi A: quoius modi B: quo vis modi C D: quoiuismodi Lorenzius (duce Guy.), Leo, Goetz-Schoell. Mala del. Camerarius. *' For en, hem, etc., with indirect questions in the subjunctive, cf . Gutsche De Inter- rogationibus OUiguis apud Ciceronem Observationes Selectae (Halle, 1885) p. 99, n. 1. 2' Becker, p. 296, would change to oriatur. 2' sit Mue. and other edd. : est codd., Tyrrell. 38 The Indicative Indirect Question in Latin Att.X12A, 2: In quo si quod Goetz-Schoell puts ? after viden, and, with Bentley, brackets ut. " vident some codd. te add. Lambinus. captat Camerarius, edd.: capiat codd. " Edd. insert non or nonne. " infert Acidalius, edd. : inferat or inferant codd. *" Becker, p. 287, would either begin a new sentence with mltu or read sit in place of e^^ " The fact that the ut-chxise intervenes between question and answer strongly favors the interpretation of that clause as an indirect question. 40 The Indicative Indirect Question in Latin Stich. 635 : Vides ut annonast gravis. Viden benignitates hominum ut periere et pro- thymiae?*' Accius, Ribbeck I p. 176, 303: Viden ut te inpietas stimulat nee moderat metus?^' Turpilius, Ribbeck II p. 98, 103: Viden ut fastidit mei?" Turpilius, Ribbeck Up. 98, 104: Ei perii! Viden ut osculatur cariem? Num hilum ilia haec pudet? Unknown Writer, Ribbeck II, p. 122, 62: Viden ut cinaedus orbem digito temperat? Afranius, Ribbeck II p. 176, 91: Viden ut facunde contra causaris patrem?^* Cic. Deorum Nat. II 44, 113: Hinc autem aspicitur Ut sese ostendens emergit Scorpios alto Posteriore trahens plexum vi corporis arcum.*' CatuU. 61, 78: Viden ut faces Splendidas quatiunt comas? CatuU. 62, 8: Viden ut perniciter exsiluere? Virg. Culex 217: Viden ut flagrantia taedis Lumina coUucent?*' Georg. I 57: Nonne vides croceos ut Tmolus odores, India mittit ebur, molles sua tura Sabaei, At Chalybes nudi ferrum, virosaque Pontus Castorea, Eliadum palmas, Epiros equarum?*^ ^ 635 vides codd., including A: viden Fl. 636 v(,ide)n A: vides P. ut perierint A' ("ut del. A''," says Lindsay): periere P, Lindsay: ut periere other edd. " mdenl ut Bamb. " viden ut Fr. Dousa: vide nunc codd. '^ viden (an video?) ut Ribbeck: videt codd.: vide ut Bentinus: iudex Buecheler. " It would be extremely forced to interpret this M<-clause as an exclamation. •' Thus Leo: lumina cum lucent cod. *»Mittet P': mittat M^C: Seneca Ep. Mor. LXXXVII 17 quotes this passage with mittat. Clauses Naturally Interpreted as Indirect Questions 41 Aen. VI 779: Viden ut geminae stant vertice cristae Et pater ipse suo superum iam signal honore?*' Group II In this group the clauses under consideration are used with an imperative or ecce. The examples in which the clauses are explanatory of pronouns, {hoc and the like) are given after the other examples. Plant. Cas. 246: Vide palliolum ut rugat. Most. 855 : Quin tu illam aspice ut placide accubat. Most. 887: Vide ut fastidit simia. True. 354: Ver vide, Ut tota floret, ut olet, ut nitide nitet."" Ter. Adel 559: Em, vide ut discidit labrum. Eun. 919: Virum bonum eccum Parmenonem incedere Video: vide ut otiosus it. CatuU. 61, 99: Vide ut faces Aureas quatiunt comas :°^ CatuU. 62, 12: Adspicite, innuptae secum ut meditata requi- runt. Laberius, Ribbeck II, p. 295, 104: Quem nulla ambitio, nulla umquam largitio, Nullus timor, vis nulla, nulla auctoritas Movere potuit in iuventa de statu: Ecce in senecta ut facile labefecit loco Viri excellentis mente clemente edita Summissa placide blandiloquens oratio! Virg. Eel. IV 52: Aspice convexo nutantem pondere mundum, Terrasque tractusque maris caelumque pro- fundum: Aspice venture laetantur ut omnia saeclo." Eel. V 7: Aspice ut antrum Silvestris raris sparsit labrusca racemis. *' st^nt many codd. and former edd. Cf. Heyne's note. '" Lindsay puts / after 354. " vide ut faces or viden? faces edd. : viden ut faces codd. " laetantur R, Wagn. Haupt.: laetentur 7y. b c GUthling Ribbeck. 42 The Indicative Indirect Question in Latin Aen. VI 856: Aspice ut insignis spoliis Marcellus opitnis Ingreditur, victorque viros supereminet omnes. Aen. VIII 192: lam primum saxis suspensam banc adspice rupem, Disiectae procul ut moles desertaque mentis Stat domus et scopuli ingentem traxere ruinam. In the following examples, the clauses under consideration fill out pronouns {hoc and the like) . Plant. Cist. 55 : Neque munda adaeque es ut soles (hoc sis vide, ut petivit Suspiritum alte) et pallida's. Cure. 126: Hoc vide ut ingurgitat inpura in se merum avari- ter faucibus plenis. Cure. 153: Hoc vide ut dormiunt pessuli pessumi Nee mea gratia commovent se ocius. Merc. 169: Ch. Hercle vero vapulabis nisi iam loquere aut hinc abis. Ac. Hoc sis vide, ut palpatur. Pseud. 152: Hoc sis vide ut alias res agunt. Aul. 47: Illuc sis vide, Ut incedit. Pseud. 955: Illuc sis vide, Ut transuorsus, non prouorsus cedit, quasi cancer solet. Ter. Adel. 229: Illud vide, Ut in ipso articulo oppressit. Eun. 670: Illud vide, os ut sibi distorsit carnufex. 30. Various Connectives. Lucr. VI 811: Nonne vides etiam terra quoque sulpur in ipsa Gignier et taetro concrescere odore bitumen; Denique ubi argenti venas aurique secuntur, Terrai penitus scrutantes abdita ferro, 810 Qualis expiret Scaptensula sub ter odores? Quidve mali fit ut exhalent aurata metalla? Quas hominum reddunt facies qualisque colores? Nonne vides audisve perire in tempore parvo Clauses Naturally Interpreted as Indirect Questions 43 Quam soleant et quam vital copia desit, 815 Quos opere in tali cohlbet vis magna necessls?*' IX. It is Incredible 4. Quae, etc. Cic. Fam. XVI 27, 2: Incredlbllest, quae ego lUos scio oppositis Gallorum castrls in aestlvis fecisse, quos ille latro, nisi allquid firmlus fuerlt, societate vitio- rum deleniet.^^ 10. Quantus, etc. Ter. Phorm. 247: Phaedrla, incredlbllest quantum erum ante eo saplentia.^^ 23. Ut. Cic. Att. IV 5, 1: Non est credibile, quae sit perfidia in istis principlbus, ut volunt esse et ut essent, si quic- quam haberent fidel.^' " The editors place / after 811 and after 812. It seems to me a little more natural to regard quidve . . . fit and quos . . . reddunt as indirect questions, parallel to quaUs expiret. " sciam Wesenberg. " est bracketed by Bentley, with comment: "Si legas incredihile est, consequens erit ut anteeam legas." Becker (p. 236) and Tyrrell foUow Bentley. " In the last three passages it would be very forced to interpret the apparent instances of indirect questions as exclamations. CHAPTER IV Indeterminate Examples: Indirect Questions or Relative Clauses I. Clauses whose Interpretation is Indifferent 1. Introductory. The relative clause is, in most of its uses,' quite distinct from the indirect question. A relative clause either modifies some particular word or words, expressed or understood, in the main clause, or else it modifies the main clause as a whole.^ An indirect question, on the other hand, is a direct object^ or a subject,* or else is in apposition with some substantive, usually itself an object or a subject.^ A relative clause is an adjectival or adverbial clause. An indirect question is a substantive clause. Frequently, however, the two distinct constructions of the indirect question and the substantive with modifying relative clause, perform practically the same function. Compare, for example, Plaut. Amph. 425, Nam quod egomet solus feci, nee quisquam alius adfuit,/In tabernaclo, id quidem hodie numquam poterit dicere, with Pseud. 696, Id tu modo me quid vis facere fac sciam. If, now, the dependent clause is introduced by a connective that is indeterminate in form — e.g., cuius, quae^ — , it may result that the clause itself is indeter- ' Just how the two kinds of clauses are related historically, is a matter of con- jecture. It is not necessary, for our purpose, to discuss the difficult problem of the origin of the Latin relative clause. For theories concerning its origin, cf., e.g., Paet- zolt, De Latini Pronominis Relativi Syntaxi Prisca 13 fi.; Delbruck, Vergl. Syntax, in, 389 fiE. (§183); KroU, Der lateinische RdaUvsatz 1 £E. * Cf . Deecke, De Usu Pronominis Relativi apiid Poetas Veteres Latinos Quaes- tiones Syntacticae (Gottingen, 1907), 54. ' E.g., Plaut. Poen 1111 : Sed earum nutrix qua sit facie, mi expedi. * E.g., Cic. Att. XV, I, 1 : Incredibilest quanta me molestia affecerit. ' E.g., Plaut. Cure. 396: Nam quid id refert mea, an aula quassa cum cinere efEossus siet?; Capt. 407: Haec pater quando sciet, Tyndare, ut fueris animatus erga suom gnatum atque se, Numquam erit tam avarus quin te gratiis emittat manu. ' There is no differentiation in form between the interrogative and the relative pronoun, except in the substantive uses of the masculine nominative singular {quis, qui) and of the neuter nominative and accusative singular {quid, quod). See Sommer, Handhuch der lateinischen Laut- und Formenlehre (Heidelberg, 1914) 434 ff.; also Appendix III of this study. Moreover, there is a large niunber of- other words which have both relative and interrogative uses (e.g., ubi, ut). 44 Clauses whose Interpretation is Indiferent 45 minate, i.e., that it may be interpreted either as relative or as inter- rogative. Cf. Most. 505: Quae hie monstra fiunt, anno vix possum eloqui. 2. Examples of Clauses whose Interpretation is Indiferent. There are numerous examples in which it is impossible to distinguish between the indirect question and the relative clause. Illustrations follow. II. Inform 4. Quae, etc. Plaut. Amph. 50: Nunc quam rem oratum hue veni, primum pro- loquar. Amph. 133: Quae illi ad legionem facta sunt, memorat pater Meus Alcumenae. Amph. 417: So. Egomet mihi non credo, quom illaec autum- are ilium audio: Hie quidem certe quae illic sunt res gestae memorat memoriter. Capt. 360: Nunc tu ilium si illo's missurus, dice monstra praecipe, Quae ad patrem vis nuntiari. Capt. 416 : Si ego item memorem quae me erga multa f ecisti bene, Nox diem adimat. Most. 505 : Quae hie monstra fiunt, anno vix possum eloqui. Poen. 556: Sed agite igitur, ut sciam uos scire rem, Expedite mihi quae vobis dudum dixi dicite. Accius, Ribbeck I p. 192, 432: Neqtte ratum est quod dicas, neque quae' agitas dicendi est locus. Ter. Adel. 604 : Sed si aliter putas, Egomet narrabo quae mihi dixti. Mi. Immo ego ibo. Hec. 362: Nequeo mearum rerum initium ullum invenire idoneum, Unde exordiar narrare quae necopinanti acci- dunt; ' agitas dicendi est locus. Ribbeck, following Voss, inserts ea before qwie. 46 The Indicative Indirect Question in Latin Partim quae perspexi hisce oculis, partim per- cepi auribus. Cic. Verr. Act. II, II 73, 179: Meminero ... me Siculis satis esse factu- rum, si quae cognovi in Sicilia, quae accept ab ipsis, diligenter exposuero. Verr. Act. II, IV, 60, 135: Longum est et non necessarium commemo- rare quae apud quosque visenda sunt tota Asia et Graecia.^ Div. I 38, 82: "Si sunt di neque ante declarant hominibus quae futura sint, aut non diligunt homines, aut quid eventurum sit ignorant, aut existumant nihil interesse hominum scire quid sit futurum, aut non censent esse suae maiestatis prae- significare hominibus quae sunt futura, aut ea ne ipsi quidem di significare possunt. . . ."• Div. II 49, lOl;!" Fam. XII 13, 3: Ibi quae cognovimus scribere ad vos quam celerrime voluimus. Sulpicius in Cic. Fam. IV 5, 4: Quae res mihi non mediocrem consolationem attulit, volo tibi commemorare, si forte eadem res tibi dolorem minuere possit.^^ Cic. Quint. Fratr. III 7, 2: Romam cum venero, quae perspexero scribam 1 ad te et maxime de dictatura. 8 sint R, Muller. " quae sint futura H. The same quotation occurs II 49, 101, where Muller has n. cr.: "sunt scr. ut p. 176, 35 (this passage) et codd. et edd. habent, sint per se non intolerabile hie utrique." '" See the critical note just above. " Mue. n. cr: ''attiderit Wesenb., Bait., Andres. Indicativum Servio condonamus.'' Kroll, Glotta III (1910-12) 6, says that the mdicative is defended by Schmalz, Z.f.g.W., 1881, p. 124. Clauses whose Interpretation is Indifferent 47 Att. X 16, 2: Tu, dum adsumus, non modo quae scies audierisve, sed etiam quae futura providebis, scribas velim.^^ Balbus, in Att. IX 7A, 2: A quo si erit nobis rescriptum, statim quae sentiemus, ad te scribemus.'' Prop. Ill 6, 1 : Die mihi de nostra, quae sentis, vera puella. Ov. Met. XIII 206: Longa referre mora est, quae consilioque manu- que Utiliter feci spatiosi tempore belli. Other examples from Ovid are Her. IX 122 and X 79; Ex Ponto III 3, 3. 13. Cur. Lucr. II 765 : Perfacile extemplo rationem reddere possis Cur ea quae nigro fuerint paulo ante colore, Marmoreo fieri possunt candore repente." 23. Vt. Plant. Amph. 599: Ordine omne, ut quicque actumst, dum apud hostis sedimus, Edissertavit.'* Bacch. 1097: Omniaque ut quidque actumst memoravit. Amph. 1129: Simul banc rem ut factast eloquar. As. 731: Nunc rem ut est eloquamur. These clauses with ut are either relative clauses, or indirect questions of the proleptic or of the pleonastic type. For pleonasm and prolepsis see Lindskog, Quaestiones, pp. 72 fiE. III. Find Out 4. Quae, etc. Ter. Phorm. 737: So. Neque ille investigatur, Ch. Quid ago? So. Qui eius pater est. Ch. Adeo, maneo, dum haec quae loquitur magis cognosco? ^ scies Ml, Wesenberg, Purser: sderis M^ Baiter, Boot, Tyrrell. ^ quae sentiemus 'NPyBa.itei, Boot, Tyrrell: cum sentiamus M^:quid sentiamtts Wesenb. " possunt codd. : possint Lambinus, edd. IS ordine omne or ordine omnem codd. 48 The Indicative Indirect Question in Latin Sail. Jug. 46, 1 : Interea lugurtha, ubi quae Metellus agehat ex nuntiis accepit, . . . diffidere suis rebus; ac turn demum veram deditionem facere conatus est. Hor. Ep. I 17, 3: Disce, docendus adhuc quae censet amiculus, ut si Caecus iter monstrare velit. Manil. II435: His animadversis debes, quae proxima cura, Noscere tutelas adiectaque numina signis, Et quae quoique deo rerum natura dicavit.^' 23. Ut. Plaut. Men. 679: Uxor rescivit rem omnem, ut factumst, ordine. If this is an example of an indirect question, it shows prolepsis. For prolepsis see Lindsay, Quaestiones, pp. 75 £E. IIIB. Observe 4. Quae, etc. Q. Cic. De Petit. 9, 35: In salutatoribus . . . hoc ef&ciendum est, ut hoc ipsum minimum ofl&cium eorum tibi gratissimum esse videatur. Qui domum tuam venient, significato te animadvertere ; eorum amicis, qui illis renuntient, ostendito, saepe ipsis dicito." 10. Quantus, etc. Varro, R. R. 1 44, 1 : Seruntur fabae modii IV in iugero . . ., sed non nullis locis paulo amplius aut minus. Si enim locus crassus, plus; si macer, minus. Quare observabis quantum in ea regione con- suetudo erit serendi, ut tantum facias; quod tantum valet regio ac genus terrae, ut ex eodem semine aliubi cum decimo redeat aliubi cum quinto decimo, ut in Etruria locis aliquot.^' " See the critical note in Garrod. " Mue. n. cr: "veniant Rob. Steph. Buechel. nescio an non necess. lis qui venient, significato te adrmadvertere eos venire ita, ut vel amicis ostendas vel ipsis dicas; . . . Post veniant add. Us dett. codd., Bait., Wesenb." " quod tantum Ursinus, Keil, Goetz: quantum is the manuscript reading. Clauses whose Interpretation is Indiferent ' 49 Keil explains the quantum-cla,use as relative: "In serendo observandum est ut tantum seras, quantum in ea regione con- suetudo est serendi." HIE. Consider, Reflect 4. Quae, etc. » Plaut. True. 931: Venitne in mentem tibi quod verbum in cavea dixit histrio? Omnes homines ad suom quaestum calent et fastidiunt. Ter. Hec. 405 : Lacrumo, quae posthac futurast vita quom in mentem venit, Solitudoque. In this example, it is doubtful whether quae posthac futurast is a relative clause modifying vita, which is in that case the subject of venit, or whether quae posthac futurast vita solitudoque is an indirect question, itself the subject of in mentem venit. Q. Cic. De Petit. 1, 3: Et saepe, quae de Demosthenis studio et exercitatione scripsit Demetrius, recordare. IV. Know 4. Quae, etc. Plaut. Cas. 668: Ly. Scelestissumum me esse credo. Par. Immo si scias dicta quae^^ dixit hodie. M. G. 281: Nescis tu fortasse apud nos facinus quod^' natumst nouom. Pacuvius, Ribbeck Ip. 132, 407: Nam si qui quae eventura sunt provideant, aequiperent lovi.^" Ter. Hec. 217: An, quia ruri crebro esse soleo, nescire arbi- tramini Quo quisque pacto hie vitam uostrarume xigat? Multo melius hie quae fiunt quam illi ubi sum adsidue scio. " For clear examples of indirect questions with this word-order, cf. Plaut. Aul. 778: Si me nouisti minus, Genere quo sim gnatus; Bacch. 891: lam dudum hercle equidem sentio, suspicio quae te sollicitet; Cure. 321: Immo si scias reliquiae quae sint. For clear examples of scire with a direct object, cf. Poen. 547: Scitis rem; Poen. 553: Scimus rem omnem. "" See Ribbeck, ad loc, for various ''emendations.'' so * The Indicative Indirect Question in Latin Cic. DeOrat. nil, 45: Ex iis enim fontibus, unde . . . dicendi praecepta sumuntur, licebit etiam laudationem ornare neque ilia elementa desiderare, quae ut nemo tradat, quis est qui nesciat quae sunt in homine laudanda?^^ Tusc. IV 36, 77: "A. Quis homo te exsuperavit usquam gen- tium impudentia? M. Quis autem malitia te? . . ." Nosti quae sequuntur. Att. XII 2, 1: Habes quae dum tu abes locuti sunt.^^ For habere followed by an indirect question, cf. the subjunc- tive examples cited by Gaffiot, "(Quis) Quid Relatif," Reme de Philologie, XXXIV (1910), pp. 73 £f. 7. Quo Modo, etc. Plaut. Bacch. 474: Ego omnem rem scio quem ad modumst.'' 23. Ut. Plaut. Pseud. 1312: Omnia ut quicque egisti ordine scio.^'* V. Hear 4. Quae, etc. Plaut. Amph. 977: Audis quae dico,^* tametsi praesens non ades. As. 447: Me. Peril hercle: iam hie me abegerit suo odio. Li. Heus iam satis tu: Audin quae loquitur? Le. Audio et quiesco. Bacch. 273: Porro etiam ausculta pugnam quam voluit dare. For indirect questions with this word-order, see p. 49, n. 19. An example which might support the interpretation that pugnam 2> sM L. ^ sunt M, vulg. : sint 2 Lambinus, Mue. ^ The dependent clause is either a relative clause or an indirect question with prolepsis. For prolepsis cf . Lindskog, QuaesHones, pp. 75 ff. 23a The Mi-clause is either a relative clause or an indirect question with pleonasm. For pleonasm cf. Lindskog, QuaesHones, pp. 72 ff. ''^With this example and the succeeding ones, compare the following relative clauses: Plaut. Cas. 165, Ecquis haec quae loquor audit?; Rud. 1129, Audi nunciam, Palaestra atque AmpeUsca hoc quod loquor; Lucilius 1032, Hoc etiam accipe quod dico, nam pertinet ad rem. Clauses whose Interpretation is Indiferent 51 is the object of 452: Immo si domum. Plaut. Bacch. 698: Bacch. 861: Bacch. 983: Capt. 313: Men. 478: Men. 909: M. G. 1222: Pseud. 194: Pseud. 230: Stich. 197: Cic. Att. X 18, 1: 4. Quae, etc. Plaut. Men. 685 : ausculta and the g'Maw-clause is relative, is Epid. audias Meas pugnas, fugias manibus dimissis Immo si audias quae dicta dixit me aduorsum tibi. Ck. Quid dixit? Audin quae loquitur? Lacrumans tacitus auscultabat quae loquebar. Est profecto deus qui quae nos gerimus audit- que et videt. Nequeo quae loquitur exaudire clanculum. Audin quae loquitur? Audin quae loquitur? Pa. Audio. Audin furcifer quae loquitur? Satin magnificus tibi videtur?^^ Pseudole, non audis quae hie loquitur? Ps. Audio, ere, equidem atque animum aduorto. Quae loquitur auscultabo, prius quam con- loquar. Itaque posthac non scribam ad te, quid facturus sim, sed quid fecerim; omnes enim KwpvKoioL videntur subauscultare quae loquor. VI. See Video quam rem agis. Quia commisi, ut me defrudes, ad eam rem ad- fectas viam. Stich. 116: An. Ubi facillume spectatiir mulier quae in- geniost bono? Pan. Quoi male faciundist potestas, quae ne id faciat temperat. For this word-order in indirect questions, see p. 49, n. 19. Ter. Eun. 783: Viden tu, Thais, quam hie rem agit? VIIB. Care Ego quae tu loquere fiocci non facio, senex. '^ quae loquitur om. A.: Lindsay brackets. 4. Quae, etc. Plaut. Rud. 782 52 The Indicative Indirect Question in Latin Ter. Hec. 513: Postremo inter se transigant ipsi, ut lubet, Quando nee gnatus neque hie mi quicquam obtemperant, Quae dico parvi pendunt. VIII. Wonder 7. Quo Modo, etc. Cic. C. M. 4, 12: Multa in eo viro praeclara cognovi; sed nihil admirabilius quam quo modo ille mortem fili tulit, clari viri et consularis. This passage is cited, by Kroll, Glotta III (1910-12), p. 6, as one in which relative clause and indirect question are indistinguish- able. II. Clauses Which, if an Indicative Indirect Question Exists, ARE More Naturally Interpreted as Indirect Questions than as Relative Clauses 1. Indications that Clauses are Indirect Questions rather than Relative Clauses. There are a number of circumstances which, in a greater or less degree, favor the interpretation of clauses as indirect questions rather than as relative clauses. (1) Sometimes the character of the introductory verb renders it unnatural to supply mentally the antecedent that will need to be understood if the clause is relative. Cf. Plant. Amph. 17: Nunc quoius iussu venio . . . dicam. Dico — unlike mitto, for example — rarely takes a personal object, and so it is less natural to supply eum or hominem than to interpret the dependent clause as an indirect question. (2) Certain relative words are not used, or not freely used, with- out an expressed antecedent. I can find no sure case of the relative adverb qui used in this way. It is natural, therefore, to interpret as interrogative the dependent clause in Plant. Trin. 165: Quid tibi ego dicam qui illius sapientiam. . . . Paene ille . . . pessum dedit? The use of the relative pronouns quo, qua, unde, ubi, and cur with- out an expressed antecedent is somewhat restricted. The first four Indications that Clauses are Indirect Questions 53 pronouns are freely so used when their antecedents are respectively eo, ea, inde, and ibi. Furthermore, there occur combinations like Hoc est unde, Hoc est cur, Habeo ubi, where the logical antecedent is id, or locus, or causa. But such free omission of the antecedent as sometimes occurs in the case of qui, quae, quod, is unheard of for these pronouns. Thus one may say, Quod hie fecit idem est atque quod ille fecit; but never. Cur hoc fecit idem est atque cur illud fecit, or Ubi ego fui idem est atque ubi tu fuisti. In view of the fact that the relatives quo, qua, unde, ubi, and cur are not freely used without expressed antecedents, it seems reasonable to interpret clauses like the following as indirect questions: Plaut. Trin. 938: Nisi quia lubet experiri quo evasurust denique; Most. 969: Scio qua me eire oportet et quo venerim novi locum; Manil. IV 117: Nee refert scelus unde cadit; Plaut. Aul. 63 : Metuo . . . Neu persentiscat aurum ubist abscondi- tum; Cic. Att. XII 40, 3: Si quis requirit cur Romae non sum (codd. vary). (3) One does not easily supply an antecedent in a case other than the nominative and accusative and other than the case of the pronoun. So in Hor. Serm. II 4, 38, Ignarum quibus est ius aptius (cited p. 68), the quibus -cldiUSQ is most probably an indirect question. Cf. also Apul. Met. V 31, 379 (cited p. 135). The ellipsis of demonstra- tive pronouns before relative clauses in examples resembling this one, though it occurs, is rare. For its occurrence cf. Bahrens, "Beitrage zur lateinischen Syntax," Philologus, Supplementband, XII (1912), pp. 324 f. Cf. also Prescott's review of Helm's edition of Apuleius, Metamorphoses, Classical Philology, X (1915), 358. (4) If the introductory verb requires an object, and the only possible objects are the indeterminate clause — which in this case must be interrogative; cf. p. 44— and some word to be supplied, the former object would seem to be the more available one. Cf. Plaut. Men. 808: Ma. Quin etiam nunc habet pallam, pater, Et spinter quod ad banc detulerat nunc, quia rescivi, refert./5e. lam ego ex hoc ut factumst scibo. It is more natural to interpret, "I shall learn from him what the facts are" (literally, "how it was done"), than, "I shall learn the facts from him, just as they occurred" (literally, "as it was done"). (5) Similarly, if there is a noun or pronoun that requires to be filled out by an appositive, it is more natural to regard the indeter- 54 The Indicative Indirect Question in Latin minate clause — which in this case must be interrogative — as the appositive, than to supply a noun and consider the indeterminate clause as relative. Cf. Plaut. Most. 1040: Eademque opera haec tibi narravero,/Quis med exemplis hodie eludificatus est. The natural interpretation is that haec has, in apposition with it, the entire dependent clause (in this case an indirect question). It would be far less natural to supply exempla as the appositive to haec, and to interpret: "I will tell you these things: (namely) the ways^in which he made sport of me." Cf. also Amph. 442: Certe edepol, quom ilium contemplo et formam cognosco meam,/Quem ad modum ego sum . . . nimi' similest mei. It might be possible, but it would not be natural, to supply modum as an appositive with formam and an antecedent of quem ad modum. (6) The occurrence, in co-ordination with an indeterminate clause, of a certain or probable example of an indirect question, seems to me to favor the interpretation of the indeterminate clause as an indirect question. Cf. Cic. Lael. 15, 55: Etenim cetera cum parant, cui parent nesciunt, nee cuius causa laborant. There do occur examples of an indirect question in co-ordination with a substantive word and modifying relative clause. Cf. Ter. Andria 356: Ausculta paucis; quid ego te velim et tu quod quaeris scies.^^ Therefore, the occurrence, in co-ordination with an indeterminate clause, of a clear example of an indirect question, does not necessitate — though, as it seems to me, it renders natural — the interpretation that the inde- terminate clause is an indirect question. (7) There are clauses which yield a better, that is, a more natural, sense if they are understood as interrogative, while yet they yield a possible sense if they are understood as relative. Cf. Plaut. Pers. 646, Quis fuit? Die nomen. Vi. Quid ilium miserum memorem qui fuit?/Nunc et ilium Miserum et me Miseram aequomst nomi- narier. It is far more natural to interpret, "Why should I tell who that poor man was? Now he should be called . . .," than "Why should I tell of that poor man who is dead?" Cf. also Plaut. Most. 459, Non potest/Dici quam indignum facinus fecisti et malum, where ™ Other examples are given by Draeger, §464. It seems to me that a like co- ordination occurs in Plaut. Rud. 965, At ego quo pacto inventust scio/Et qui invenit hominem novi et dominus qui nunc est scio. However, Lindskog, Qttaestiones, 73, sees in qui invenit hominem novi an indirect question with pleonasm. Indications that Clauses are Indirect Questions 55 the gMaw-clause yields a more natural sense if it is understood as interrogative than if it is understood as relative.^' There are other examples in which there is less difference between the two interpreta- tions, but in which the interpretation that the dependent clause is interrogative )iields a somewhat more natural sense. So for Plant. Amph. 460, Ibo ad portum atque haec uti sunt facta ero dicam meo, the interpretation, "I will tell how these things were done" is some- what more natural than, "I will tell these things just as they were done." In the following example it makes still less difierence which of the two interpretations is adopted: Cic. C. M. 6, 18, At senatui quae sunt gerenda praescribo. (8) A clause introduced by (quis) quid is, when not certainly interrogative, at least more likely to be interrogative than relative. The use of relative (quis) quid is much more rare than the use of " Becker argues from the existence of clauses of the type of Ter. Hec. 417 (cited below) that clauses of the type of Most. 459 are relative clauses with tarn or tantum understood (pp. 308-9; unfavorably criticized by Kroll, p. 5). While it is conceivable that the clauses of the type of Most. 459 are relative, it seems to me that they are far more probably interrogative. I know of only four examples of the type of Hec. 417. They are: EnniuSjRibbeck p. 62,324: Non potis ecfari tantum dictis, quantum factis sup- petit. [non potis ecfari Davisius: non potest ecfari Brix: non potest haecfari Gud. Reg. tantum om. Bentley.] Ter. Hec. 417: Non hercle verbis, Parmeno, did potest Tantum quam reapse navigare incommodimist. [quantum iorguam A. reapse Tyrrell: re ipsa codd.] Cic. Att. VI 4, 3: Non queo tantum, quantum vereor, scribere. [Shuckburgh (London, 1908-09) translates: "I cannot express the extent of my fear."] Buecheler 709, 12 (CH. V 6728; an inscription in honor of Flavianus, a bishop who is thought to have died in 542) : Tantaque fari nequeo, quanta insunt gratiae opes. It is to be noted that in every one of these instances the main clause states that it is impossible to express some idea. In other words, the qmntum-claxi&e depends on a verb of saying which, in its turn, depends upon a phrase that expresses impossibility. The following example is a clause of a different type, for indica means, "set a price" : Plant., Pars. 661: Dor. Turn tu pauca in verba confer: qui datur tanti indica. Sag. Faciamitauttevelle video, ut emas. Habe centum minis. [qui Scioppius, edd., for quid.] 56 The Indicative Indirect Question in Latin interrogative {quis) quid. For relative (gww) gMJi see Appendix III; -especially, for the classification of the examples in this study, p. 200. 2. Examples of Clauses More Naturally Interpreted as Indirect Questions than as Relative Clauses.^^ * ' I. Ask 4 and 5. Quae 6* Qui, (Masculine Singular Substantive).^' Plant. Pers. 636 and 638: Do. Ubi tu nata es? Do. At ego patriam te rogo quae sit tua. Vi. Quae mihi sit nisi haec ubi nunc sum? Do. At ego illam quaero quae fuit.^" Vi. Omne ego pro nilo esse duco quod fuit, quando fuit: Tanquam hominem, quando animam ecflavit, quid eum quaeras qui fuit? Quae fuit is most naturally understood, it seems to me, in the same way as quae sit in the preceding verse: namely, as an indirect question. In 638, the interpretation, "Why do you ask about hiin, — what sort of man he was?" is more natural than, "Why do you ask for him who is dead?" 13. Cur. Cic. Att. XII 40, 3: Si quis requirit cur Romae non sum: quia discessus est; cur non sim in eis meis praedioli?, quae sunt huius temporis: quia frequentiam illam non facile ferrem.^^ 2' The examples in this section vary greatly in the degree in which the inter- pretation as indirect questions is the more natural one. " For qui used like quis, cf. O. Sej^ert, Review of Neue, "Formenlehre der la- teinischen Sprache," B. P. W., XIII (1893), 277 fE. '" 636 qui fuit A (from vs. 638). '^cur Romae non sim edd. Baiter-Kayser n.: "Romae non sum W-. cur non sumus in eis M'l cur non sim in Ms meis M^." Clauses Naturally Interpreted as Indirect Questions 57 Varro, Reliquorum de Grammatica Librorum Fragm., Goetz-Schoell, p. 230 (Keil I 439, 15) : Nee ipsorum tamen, si interrogentur cur id secuti sunt, scientium.''' 21. Unde. Cic. Deorum Nat. 11 6, 18: Quin et umorem et calorem, qui est fusus in corpore . . ., si quis quaerat unde habemus, apparet, quod aliud a terra sumpsimus, aliud ab umore.'' , ^ . J II. Inpoem 1. Quid, Plaut. Cist. 57: Eloquere utrumque nobis, Et quid tibist et quid velis nostram operam, ut nos sciamus. It seems to me extremely improbable that this example or the ones with quid which follow are relative clauses. But see Appen- dix, p. 200. Plaut Most. 572: Da. Quin tu istas mittas tricas? Tr. Quin quid vis, cedo. The question intervenes between cedo and its modifier quin. Hence it is not a direct question. Plaut. Pseud. 696a: Id tu modo me quid vis facerefac sciam. Since the clause breaks into the main sentence, Becker's interpretation, p. 144, that it is a direct question is impossible. Plaut. Pseud. 696c: Tu modo quid me facere vis fac ut sciam.^" 4. Quae, etc. Plaut. Amph. 17: Nunc quoius iussu venio et quam ob rem venerim, Dicam simulque ipse eloquar nomen meum. Becker, p. 310, and Gaffiot, Rev. de phil., XXVIII, p. 51, inter- pret the clause as relative. This interpretation is the less '^sunt (or essent) codd.: sunt Keil: sint Putschius, Goetz-Schoell. " habemus A B C P V O (according to Halm-Baiter) : habeanms E, edd. » 696 b c cm. A, Ritschl, Leo, Goetz-Schoell. 58 The Indicative Indirect Question in Latin natural one, even apart from the parallel subjunctive clause. Cf. p. 52, criterion 1. Plaut. Amph. 45 : Nam quid ego memorem . . . . . . quis benefactis mens pater, Deorum regnator, architectus omnibus?'* Most. 1040: Eadem opera haec tibi narravero, Quis med exemplis hodie eludificatus est. Becker, p. 245, is troubled by this example and remarks: "Certa medela nondum excogitata est." To interpret, "I will tell you these things, namely, the ways in which he made sport of me," would certainly be forced. As Becker himself says. I.e., "de pronomine relativo vix ac ne vix quidem cogitari potest." Plaut. Pseud. 21: Ps. Face me certum quid tibist: . . . Cal. Cape has tabellas, tute hinc narrato tibi Quae me miseria et cura contabefacit. Rud. 1023 : Quo argumento socius non sum et fur sum facdum ex te sciam. Pomponius Bono- niensis, Ribbeck II p. 251, 175: Sed qui utrosque error vos agitat, expedibo. AdHeren, III7, 13: Divisione hac utemur: Exponemus quas res laudaturi sumus aut vituperaturi.'^ AdHeren II 31, 50: Misericordia commovebitur auditoribus, .... si ostendemus, in quibus commodis fuimus quibusque incommodis simus, comparatione.'* VarroL.L. VI5,39: Democritus, Epicurus, item alii qui infinita principia dixerunt, quae unde sint non dicunt, sed cuius modi sint, tamen faciunt magnum: quae ex [h]is constant in mundo, ostendunt. Caesar B. G. 120, 6: Dumnorigem ad se vocat, fratrem adhibet; quae in eo reprehendit ostendit, quae ipse intel- legat, quae civitas queratur proponit.'' ^ t add. Pareus. Leo thinks a line is lost. ^ Thus Marx: sinms, MOUer, with n.: "sumus H B b ft cf. Langen." Langen, Philologus, XXXVII, 406: ''sane plurimi Ubri sumus habent, sed verum Parisinus." ^in quibus commodis fuimus b, 1, d, C: om. H P II B: fuerimus Ernest, Mue., Marx. " reprehendit p (the class of codd. which is free from corrections of grammarians; cf. Kuebler, pp. VIII flE.), M: reprehendat other codd., edd. Clauses Naturally Interpreted as Indirect Questions 59 Cic. Fam. VIII 1, 2: Si quid in re publica maius actum erit, quod isti operarii minus commode persequi possint, et quem ad modum actum sit et quae existimatio secuta quaeque de eo spes est diligenter tibi perscribemus.^^ Leg. Agr. II 18, 49: Hie mihi parumper mentes vestras, Quirites, commovere videor, dum patefacio vobis quas isti penitus abstrusas insidias se posuisse arbi- trantur contra Cn. Pompei dignitatem.'^ This example is cited by Kroll, p. 6, as one in which indirect question and relative clause are indistinguishable. It seems to me most natural to interpret it as an indirect question. Lambinus and the editors who adopt his "emendation" clearly share this feeling. Cic. C. M. 6, 18: At senatui quae sunt gerenda praescribo.^" Sail. Cat. 51,4: Magna mihi copia est memorandi, patres conscripti, qui reges atque populi ira aut miseri- cordia impulsi male consuluerunt.^' Virg. Georg. II 122: Quid tibi odorato referam sudantia ligno Balsamaque et bacas semper frondentis acanthi? Quid nemora Aethiopum molli canentia lana? Velleraque ut foliis depectant tenuia Seres? Aut quos Oceano propior gerit India lucos, Extremi sinus orbis, ubi aera vincere summum Arboris haud uUae iactu potuere sagittae? Georg. IV ISO: Nunc age, naturas apibus quas luppiter ipse Addidit, expediam, pro qua mercede canoros Curetum sonitus crepitantiaque aera secutae Dictaeo caeli regem pavere sub antro. In the last two passages, Jahn and other editors interpret the quae-cla,nses as relative. It seems to me more natural to regard them as indirect questions. Lucian Miiller, on Horace Serm. ^'estMH: sit edd. " arbitrentur Lambinus, edd. *" sunt B S: sint other codd., edd. " Codd. vary. See Dietsch ad loc. 60 Tke Indicative Indirect Question in Latin II 4, 39 cites Georg. IV 150 as an instance of the indicative indirect question. Virg. Aen. VI 615: Inclusi poenam expectant. Ne quaere doceri, Qu'am poenam, aut quae forma viros fortunave mersit. Gaffiot, Rev. de Phil., XXVIII, p. 52, n. 2, interprets the dependent clause as relative. This interpretation seems to me very unnatural. Sidgwick, ad loc, remarks that "this inter- pretation would be so harsh as to be impossible." He adopts a change to merset, which, he says, is approved by Madvig. Prop. II 34, 78: Tu canis Ascraei veteris praecepta poetae, Quo seges in campo, quo viret uva iugo. "You sing of the field in which ..." (relative clause) is a less natural interpretation than "Your song tells in what field ..." (indirect question). Ovid Trist. IV, X 44: Saepe suas volucres legit mihi grandior aevo, Quaeque nocet serpens, quae iuvat herba, Macer.*2 Vitruv. II 1, 7: Igitur de his rebus quae sunt in aedificiis ad usum idoneae quibusque sunt qualitatibus et quas habeant virtutes, ut potero, dicam. Vitruv. II 8, 16: Sed id genus quid ita a populo Romano in urbe fieri non oporteat, exponam, quaeque sunt eius rei causae et rationes non praetermittam. Vitruv. IV Intro- duction, 2: Vitruv. IX 4, 6: Tertio autem [sc. disputavi] de aedium sacrarum dispositionibus et de earum generum varietate quasque et quot habeant species earumque quae sunt in singulis generibus dis- tributiones. Quae sunt ad dextram orientis inter zonam signorum et septentrionum sidera in caelo dis- posita, dixi [esse]; nunc explicabo quae ad sinis- tram orientis meridianisque partibus ab natura sunt distributa. *> nocens H P: tiecet A and other codd. imet A H and other codd. Clauses Naturally Interpreted as Indirect Questions 61 5. Qui. (Masculine Singular Substantive; cf. p. 56, n. 29.) Plaut. Pers. 646: Quisfuit? die nomen. Vi. Quid ilium miserum memorem qui fuit? Nunc et ilium Miserum et me Miseram aequomst nominarier. 6. Qua Causa, etc. Plaut. Cist. 82: Hoc volo agatis: qua accersitae causa ad me estis, eloquar. If Deecke, De Usu Pronominis Relativi, p. 54, is right, the fact that qua and causa are separated shows that the qua causa clause is not relative but interrogative. In that case, the clause is a practically certain instance of an indirect question; for to interpret it as a direct question would be so forced as to be practically impossible. Cic. Leg. Agr. II 4, 15: Et quoniam, qua de causa et quorum causa ille hoc promulgavit, ostendi, doceat ipse nunc, ego quern possessorem defendam, cum agrariae legi resisto.^' 7. Quo Mode, etc. Plaut. Trin. 578: Sed, Stasime, abi hue ad meam sororem ad Calliclem: Die hoe negoti quo modo actumst. St. Ibitur. If this clause is relative, the meaning is, "Die hoc negoti eo modo quo actumst": "Tell her this business in the way in which it was done." Such an interpretation seems to me unnatural. Cic. De Leg. 1 9, 27 : Nam et oculi nimis arguti, quem ad modum animo affecti sumus, loquuntur, et is qui appellatur vultus . . . indicat mores. . . .^ Ad Heren. IV 12, 17: Haec qua ratione vitare possumus, in arte grammatica dilucide dicemus.^^ '^ Thus codd. : promulgaverit or promulgarit Mue. and other edd. **simis H, Mue., other edd.: sumus AB (the best codd. according to Mue., p. xxxii), Hahn-Baiter. *> possumus H P B, Marx: possimus n C b 1 d, Mue. 62 The Indicative Indirect Question in Latin Ad Heren. IV 56, 69: Deraonstratum est enim, quomodo res in omnibus generibus causarum invenire oporteat; dictum est, quo pacto eas disponere convenit; traditum est, qua ratione esset pronuntiandum; praeceptum est, qua meminisse possemus; demonstratum est, quibus modis perfecta elo- cutio compararetur.'*' 10. Quantus, etc. Cic. Tusc. V41, 121:" In quo quantum ceteris profuturi sumus non facile dixerim.*' It would be very forced to interpret the quantum-c\a,Vise as relative. For the possibility that it is relative, cf. p. 55, n. 27. 11. Quot. Ad Heren. 1 10, 17 : Enumeratione utemur, cum dicemus numero, quot de rebus dicturi sumus. . . . Expositio est, cum res, quibus de rebus dicturi sumus, exponimus breviter et absolute.** Ovid Ibis 199: Nam neque quot flores Sicula nascuntur in Hybla Quotve ferat dicam terra Cilissa crocos, Nee, cum tristis hiemps aquilonis inhorruit alis, Quam multa fiat grandine canus Athos.*' 14. Qui (Adverb). Plaut. Trin. 165: Quid tibi ego dicam, qui illius sapientiam Et meam fidelitatem et celata omnia Paene ille ignauos funditus pessum dedit?*" Lucr. V 274: Aera nunc igitur dicam qui corpore toto Innumerabiliter privas mutatur in horas. Semper enim, quodcumque fluit de rebus, id omne Aeris in magnum fertur mare. r '"convenit HPB: conveniafE. The passage is bracketed by some edd., but not by Marx. *' sumus C : simus Beroald, edd. *' sumus codd. (except 0), Marx: simus /3, Muller, Langen in Phil., XXXVII, 405. " Nascuntur F G H M P T X Farm. : pascuntur Vat. : nascantur V edd. 202 fiat V. SI Becker, p. 245, suggests a change of qui to quin. Clauses Naturally Interpreted as Indirect Questions 63 I agree with M. Patin and L. von Knebel in considering this clause an indirect question. Patin (Paris, 1876) translates: "Quant a I'air, il faut maintenant que je dise quels changements innombrables ont lieu"; von Knebel (Leipzig, 1831): "Nunmehr komm ich zur Luft, wie diese zu jeglicher Stunde, Fast unzaeh- lige Mai im ganzen Bestande sich aendert." H. A. J. Munro (ed. 4, Cambridge, 1886) and C. Bailey (Oxford, 1910) take qui as the relative, meaning "which." 19. Quam. Plant. Most. 459: Non potest Dici quam indignum facinus fecisti et malum. Ter. Hec. 91: Non dici potest Quam cupida eram hue redeundi. Hec. 646: Nosne hoc celatos tam diu! Nequeo satis Quam hoc mihi videtur factum prave, proloqui. For the possibility, the bare possibility, that these three quam- clauses are relative, see p. 55, n. 27. For Hec. 646 this inter- pretation is particularly dif&cult, since satis makes it the more unnatural to supply, in thought, a tam or tantum. 23. Ut. The following clauses may conceivably be relative, as the Mi-clause in Cic. Pro Milone 10, 29 almost certainly is: Dicam enim aperte, non derivandi criminis causa, sed ut factum est. However, it is more natural to interpret them as indirect questions. Plant. Amph. 460: Ibo ad portum atque haec uti sunt facta ero dicam meo. Amph. 559: Tamen, quin loquar haec uti facta sunt hie, Numquam ullo modo me potes deterrere. Amph. 1042 : lam ad regem recta me ducam resque ut factast eloquar. As. 367 : Nunc tu abi ad forum ad erum et narra haec ut nos acturi sumus. As. 376: Le. Pugno malam si tibi percussero, Mox quom Sauream imitabor, caveto ne sus- censeas. Li. Hercle vero tu cavebis ne me attingas, si sapis, 64 The Indicative Indirect Question in Latin ■Ne hodie malo cum auspicio nomen commu- taveris. Le. Quaeso, aequo animo patitor. Li. Patitor tu item quom ego te referiam. Le. Dico ut usust fieri. Li. Dice hercle ego" quoque ut facturus sum. Bacch. 1063: Ch. Non equidem capiam. Ni. At quaeso. Ch. Dico ut res se habet. Ni. Morare. Ch. Nolo, inquam, aurum con- credi mihi. Merc. 351: Nunc si dico ut res est atque illam mihi me Emisse indico, quern ad modum existumet me? Rud. 1211: Da. Eloquere ut haec res optigit de filia; Eum roga ut relinquat alias et hue veniat. Trin. 712: Nil ego in occulto agere soleo: mens ut animust eloquar. si sententiast. Trin. 749: Ipsum t adeam Lesbonicum edoceam ut res se habet."* Ter. Adel. 513: Ego Micionem, si apud forumst, Conveniam atque ut res gestast narrabo ordine. Virg. Aen. VII 207 : Atque equidem memini (fama est obscurior annis) Auruncos ita ferre senes, his ortus ut agris Dardanus Idaeas Phrygiae penetravit ad urbes.^^ 30. Various Connectives. Sail. Cat. 23, 5: At Fulvia, insolentiae Curi caussa cognita, tale periculum rei publicae haut occultum habuit, set sublato auctore de Catilinae coniura- tione, quae quoque modo audierat, compluribus narravit.^' -* „ ^ IIB. Discuss 0. Qua Causa, etc. Ter. Phorm. 798: Quid tu? Ecquid locutu's cum istac quam ob rem banc ducimus? '' usus codd.: usust Lambinus, edd. ^^'^habeatC '^penetravit codd. except R; note in Ribbeck: "peneirarit R, Bentley, Hirtzel." 6' See Dietsch ad he. Clauses Naturally Interpreted as Indirect Questions 65 lie. Depict 7. Quo Modo, etc. Ter. Eun. 585: Ibi inerat pictura haec, lovem Quo pacto Danaae misisse aiunt quondam in gremium imbrem aureum. III. Find Out 4. Quae, etc. Cic. Att. XIII 30, 3 : Mi sicunde potes, erues qui decern legati Mummio fuerunt.^^ Lentulus, in Cic. Fam. XII 15, 6: Si percurrero provinciam et cognovero qui nobis et rei p. fidem praesPiterunt in conservanda pecunia a me deposita, quique scelere ultro deferentes pecuniam publicam hoc munere societatem facinorum cum Dolabella inierunt faciam vos certiores.^^ 7. Quo Modo, etc. Cic. Verr. Act. II, II 53, 131: lam vero censores quem ad modum isto praetore in Sicilia creati sunt, operae pretium est cognoscere.^^ 20. Ubi. Plaut. Aul. 63 : Nimisque ego banc metuo male, Ne mihi ex insidiis verba inprudenti duit Neu persentiscat aurum ubist absconditum. Becker's interpretation (p. 311) that the clause is relative, with locum understood, seems lo me very forced. 23. Ut. Plaut. Men. 808: Ma. Quin etiam nunc habet pallam, pater, spinter, quod ad banc detulerat, nunc, quia rescivi, refert. Se. lam ego ex hoc ut factumst, scibo. Becker's interpretation (p. 307) that the clause is relative is, it seems to me, quite unnatural. Cf. p. 53, criterion 4. '^fuerint edd. Bai.-Kay. has n. cr.: "fuerint ed. Romana; /««■«>«« M." M praestiterint . . . inierint Wesenb., Bai. ™ sunt Lg. 29: sf Lg. 42: sint other codd., Lambinus, edd. 66 The Indicative Indirect Question in Latin IIIB. Observe 4. Quae, etc. Plaut. Capt. 384: Most. 399: 4. Quae, etc. Livy XLI 24, 4: 4. Quae, etc. Plaut. Poen. 1185: Ergo animum aduortas uolo, Quae nuntiare hinc te uolo in patriam ad patrem. Animum aduorte nunciam tu, quae uolo ac- curarier. HID. Determine, Judge Divinat etiam, quae futura fuerant si Philippus vixisset: quid ita Perseus regni haeres sit, quid parent Macedones, quid cogitent Romani.'' HIE. Consider, Reelect Ante. Spero equidem. Ade. At pol ego, quom ingeniis quibu' sumus atque aliae gnosco.^' Ad Heren. 20, 31: II Cic. Fin. IV 24, 67: Quoniam igitur ostendimus perfectam et planam argumentationem ex quinque partibus constare, in una quaque parte argumentationis quae vitia vitanda sunt consideremus.^^ Vestri autem progressionem ad virtutem fieri aiunt, levationem vitiorum fieri negant. At quo utuntur homines acuti argumento ad probandum, operae pretium est considerare.*" Ovid Met. XII 473 : Et te, Caeni, feram? nam tu mihi femina semper, Tu mihi Caenis eris, nee te natalis origo Commonuit, mentemque subit, quo praemia facto Quaque viri falsam speciem mercede parasti?^'- "fuerint many edd. '' ig(no)s(co) A. ^' sint Mue. with n. cr.: "sunt codices, sint editores." Langen, Phil., XXXVII, 406: ''sunt legitur in prima famUia nonnullisque aliis." Langen would read sint. Marx reads sunt. 6» utantur Lambinus, Mue., and other edd. '^ pararis N. Clauses Naturally Interpreted as Indirect Questions 67 7. Quo Modo, etc. Plaut. Amph. 442 : Certe edepol, quom ilium contemplo et formam cognosce meam, Quem ad modum ego sum . . . nimi' similest mei. IV. Know 1. Quid. Plaut. True. 786: Nee quid peccavi scio.*^ It is very improbable that the quid-c\a,\ise is relative. But see Appendix, p. 200. 4. Quae, etc. Plaut. Capt. 206b: Quid a nobis metuit? Scimus nos Nostrum officium quod est, si solutos sinat. For this word-order in indirect questions cf. p. 49, n. 19. For the use of quod in place of quid, cf. M. G. 1344, Sed quid hoc? Quae res? Quod video? {quod cod., quid Ital., edd.) ; Poen. 547, Narravi uobis quod nostra opera mi opus siet {quid Ussing); Publ. Syr., Ribbeck II, p. 341, 375: Minus saepe pecces, si scias quod nescias {quid editors before Wofflin); Cic. Fam. Ill II, 2: Quod enim?; Att. IV 7, 3: Quod superest? Plaut. Capt. 207: At fugam fingitis: sentio quam rem agitis. Pers. 109: Sed ecquid meministin here Qua de re ego tecum mentionem feceram? Sat. Memini: ut murena et conger ne cale- fierent. Pers. 515: Nescis quid te instet boni Neque quam tibi Fortuna faculam lucriferam adlucere uolt. GafEot, Rev. de Phil., XXVIII 51, interprets the clause as relative. This interpretation seems to me extremely forced. Becker, p. 220, would emend. Plaut. Rud. 958: Furtum ego vidi qui faciebat; Noram dominum, id cui fiebat. Post ad furem egomet devenio •2 nee or net codd. Becker (p. 262) suggests a change either of quid to quod or of peccavi to peccarim. Leo and Lindsay read quod. Goetz-Schoell reads quid, but suspects the passage. 68 The Indicative Indirect Question in Latin Feroque ei condicionem hoc pacto: 'Ego istuc furtum scio quoi factumst: Nunc mihi si vis dare dimidium, ludicium domino non faciam.' Becker, p. 310, interprets this clause as relative. However, so far as I can find, scire has nowhere in early or classical Latin a personal object (unless there is prolepsis, as in Plant. Men. 301, Neque te qui homo sis scio.). Ter. Adel. 309: Non intellego Satis quae loquitur. Ca. Propius obsecro ac- cedamus, Sostrata.'' It would be unnatural to interpret: "I do not understand the things which he is saying." Propius obsecro accedamus shows that the defect is not one of imperfect comprehension, but of imperfect hearing. With this meaning of intellego the interrogative construction harmonizes better than the relative. Cic. Att. IX 13, 3: Nee ego nunc, eum iuvare qua re possum, scio nee, si possem . . . adiuvandum putarem.'* Fam. II 9, 1: Itaque, cum primum audivi, ego ille ipse factus sum — scis quem dico — egique omnis illos adulescentes, quos ille actitat.*^ Cf. the remark on Rud. 958, above. Lael. 15, 55: Etenim cetera cum parant, cui parent nesciunt, nee cuius causa laborant . . . ami- citiarum sua cuique permanet stabilis et certa possessio.^' Hor. Serm. II 4, 38: Nee satis est cara piscis averrere mensa Ignarum quibus est ius aptius et quibus assis Languidus in cubitum iam se conviva reponet. For the possibility, the bare possibility, that this clause is relative, cf. p. 53, criterion 3. It seems to me almost certainly interrogative, and indeed the editors generally consider it an indirect question. Cf. the comments of Lucian Miiller (1891), " satius quae loquitur A: satis quae loquatur Pi? S. •* possum M : possim edd. "* dicam Mendelssohn, Bai.-Kay., Ernest, Tyrrell-Purser, without n. cr. Reisig- Schmalz, p. 494, says manuscript-reading is dico. ™ laborantur P: laborant the other codd: labor ent edd. Clauses Naturally Interpreted as Indirect Questions 69 Palmer (1891), Lejay (1911). Heindorf alone (Breslau, 1815), so far as I know, suggests that the clause may be relative: "Ignarum, sc. eorum, quibus. Oder est und reponet stehen hier nach antiker und griech. Art fiir den Conjunktiv." Manil. IV 211: Et licitum sciet et vetitum quae poena seque- tur." 5. Qui (Masculine Singular Substantive; cf. p. 56, n. 29). Plaut. Rud. 965 : Tr. Vidulum istum quoius est novi ego hominem iam pridem. Gr. Quid est? Tr. Et quo pacto periit. Gr. At ego quo pacto inventumst scio: Et qui invenit hominem novi, et dominus qui nunc est scio. Nilo pol pluris tua hoc quam quanti illud refert mea. Ego ilium novi quoius nunc est: tu ilium quoius antehac fuit. Rud. 1297: Meum hercle illic homo vidulum scit qui habet, ut ego opinor. For the last two examples, cf. Rud. 958, p. 67. 6. Qua Causa, etc. Plaut. Men. 715 : Non tu scis, mulier, Hecubam quapropter canem Graii esse praedicabant? Ma. Non equidem scio. Cic. Fin. V 9, 24: Tum sensim incipit progredi seseque agno- scere et intellegere, quam ob causam habet eum, quem diximus, animi appetitum.^^ 7. Quo Modo, etc. Plaut. Aul. 47 : At scin quo modo tibi res se habet? ■ M. G. 257: Atque haec ei dice, monstra, praecipe, Ut teneat consilia nostra, quem ad modum exorsi sumus, De gemina sorore. *' sequetur M 1? Van Wageningen: sequentur V: sequatur G. " habet codd., Gaffiot, Pour le Vrai Latin 25: habeal Lambinus; edd. (note from Halm-Baiter) : Mue. reads hdbeat, without note. 70 The Indicative Indirect Question in Latin Ter. Adel. 636: Ego Aeschinem conveniam, ut quo modo acta haec sunt sciat.'^ Plaut. Rud. 964: Tr. Vidulum istum quoius est novi ego hominem iam pridem. Gr. Quid est? Tr. Et quo pacto periit. Gr. At ego quo pacto inventust scio. Becker (p. 312) interprets these clauses as relative. 13. Cur. Cic.Verr.Actll, III 26, 64: lam omnes intelligunt cur universa pro- vincia defensorem suae salutis eum quaesivit, cuius iste fidei, diligentiae, perseverantiae nulla ratione eripi possit.'"' 16. Qua. Plaut. Most. 969: Scio qua me eire oportet et quo venerim novi locum. Gaffiot, Rev. de Phil., XXVIII, 54, interprets the qua-dause as relative. 21. Unde. Ovid Her. XV 4: Ecquid ut aspecta est studiosae littera dextrae, Protinus est oculis cognita nostra tuis? An nisi legisses auctoris nomina Sapphus, Hoc breve nescires unde movetur opus?'^ Ovid Ex Ponto I 8, 25: , Teque, quod et praesto est — quid enim tibi plenius optem? Martia cum magno Caesare Roma probet. Sed memor unde abii, queror, o iocunde sodalis, Accedant nostris saeva quod arma malis. For the possibility that the unde-c\a.\ise is relative, cf. p. 52, criterion 2. " sunt A, corr. A*; sunt B C D E F G P (critical note from Becker, p. 308) : sint Umpfenbach, Tyrrell. "> Recent edd. read quaesiverit. Earlier edd. seem to have read quaesivit; see the quotation in Ramshom, "Lateinische Grammatik" (Leipzig, 1830) II 712. Halm- Baiter has n. cr.: "quaesiverit Lg 42 H; quaesierit Lg 29 A B: quaesivit dtt." " movetur codd. (except F and m) and most edd.: veniret F, Sarravianus, Palmer: nominetur (with gloss veniret)m. according to De Vries (ed. Leyden, 1885). Clauses Naturally Interpreted as Indirect Questions 71 V. Hear 4. Quae, etc. Plaut. Pseud. 153: Hue adhibete auris quae ego loquor, plagigera genera hominum."'' Ter. H. T. 731: Mea Phrygia, audisti modo iste homo quam villain demonstravit Charini? Ph. Audivi. Ba. Proximam esse huic fundo ad dextram? Ph. Memini. Becker, p. 285, suggests that the quam-c[a.use may be relative, depending upon the following infinitive phrase, "villam Charini proximam esse huic fundo ad dextram." This does not seem to me to be the natural interpretation. Lucr. I 269: Nequa forte tamen coeptes diffidere dictis. Quod nequeunt oculis rerum primordia cerni, Accipe praeterea quae corpora tute necesse5< Confiteare esse in rebus nee posse videri. Prop. IV 8, 74: Atque ait"admissae si vis me ignoscere culpae Accipe quae nostrae formula legis erit " 7. Quo Modo, etc. Plaut. Rud. 356 : Non audivisti, amabo, Quo pacto leno claneulum nos hine auf erre uoluit In Siciliam et quicquid domi fuit in navem inposvit? Becker, p. 288, is troubled by this example: "Vixpro relative habendum est." VI. See 4. Quae, etc. Cic. Fam. XIV 17: Nunc quae sunt negotia vides.'' Rose. Am. 34, 95 : Videamus nunc strictim, sicut cetera, quae post mortem Sex. Rosci abs te, T. Rosci, facta sunt.'^ " loguor A, Leo, Lindsay: loquar P, Goetz-Schoell. " Thus codd., Gaffiot, Pour le wai laUn, p. 25, Lambinus: sint edd. " sint many edd. 72 The Indicative Indirect Question in Latin Inv. II 12, 39: Videre igitur primum oportebit, quae sint continentia cum ipso negotio, hoc est, quae ab re separari non possunt.'^ Virg. Aen. IX 269: Vidisti quo Turnus equo, quibus ibat in armis Aureus; ipsum ilium, clipeum cristasque rubentes Excipiam sorti, iam nunc tua praemia, Nise. Ovid Ars. Am. II 529: Dux bonus huic centum commisit vite regendos, Huic equites, illi signa tuenda dedit: Vos quoque, de nobis quern quisque erit aptus ad usum Inspicite et certo ponite quemque loco. Ars. Am. Ill 115: Simplicitas rudis ante fuit: Nunc aurea Roma est Et domiti magnas possidet orbis opes. Aspice quae nunc sunt capitolia, quaeque fuerunt: Alterius dices ilia fuisse lovis. In the last two examples an interpretation of the clauses under consideration as anything else than indirect questions would be very forced. 5. Qui (Masculine Singular Substantive; cf. p. 56, n. 29.). Plant. Men. 349: Videamus qui hinc egreditur.^^ Rud. 956: Furtum ego vidi qui faciebat: Noram dominum id cui fiebat. The cMJ-clause is undoubtedly relative. The gMJ-clause, however, it seems to me, is more naturally interpreted as an indirect question. 7. Quo Mode, etc. Varro, L. L. X 3, 58: Si multitudinis rectus casus forte figura cor- rupta erit, . . . prius id corrigemus quam inde ordiemur: ab obliquis adsumere oportet figuras eas quae non erunt ambiguae, sive singulares sive multitudinis, ex quibus id, cuius modi dehent esse, perspici possi[n]t.'*°' " possunt R, first hand of P, Ernest: possint A T, second hand of P, most edd. " hinc crediinir codd. (corr. B'). "a debeant changed by same hand to debent. Clauses Naturally Interpreted as Indirect Questions 73 Because of the presence of id, it would be extremely harsh to regard the cuius modi clause as relative. Cf. p. 53, criterion 5. 23. Ut. Ovid Trist. I IX 9: Aspicis, ut veniant ad Candida tecta columbae, Accipiat nuUas sordida turris aves; 9 Horrea formicae tendunt ad inania numquam; NuUus ad amissas ihit amicus opes; 1 1 UtqMe. comes radios per solis euntibus umbra est, 12 Cum latet hie pressus nubibus, ilia fugit? Mobile sic sequitur Fortunae lumina vulgus: Quae simul inducta nocte teguntur, abit. On the punctuation here adopted, tendunt, ihit, est, and fugit are instances of the indicative in indirect questions. Merkel and Owen punctuate so that they do not have any instances of the indicative indirect question. 30. Various Connectives. Ov. Trist. I III 52: A quotiens aliquo dixi properante, "Quid urges? Vel quo festinas ire, vel unde vide."" VII. Concern 4. Quae, etc. Lucr. II 814: Nee refert ea quae tangas quo forte colore Praedita sunt.'' 15. Quo. Varro R. R. I 4, 4: Quod permagni interest, ubi sint positae villae, quantae sint, quo spectant porticibus, ostiis ac fenestris.^' 21. Unde. Manil. IV 117: Nee refert seelus unde cadit; seelus esse faten- dum.*" "fesUnes dett. " sint Aid., Junt., other edd., for sunt. Tangas is in the subjunctive because it is an instance of the indefinite second person singular in a generalizing clause. Cf. Hale-Buck, §504, 2. " spectant P A B : spectent v, edd. 8» cadat g, many add.; cadit the other codd., Van Wageningen. 74 The Indicative Indirect Question in Latin VIII. Wonder 7. Quo Modo, etc. Laberius, Ribbeck II, p. 282, 20: Mirabar quo modo mammae mihi Tam descendiderant. X. No Verb Expressed 7. Quo Modo, etc. Varro, L. L. VIII 1, 1: Quom oratio natura tripertita esset . . ., cuius prima pars, quemadmodum vocabula rebus assent imposita, secunda, quo pacto de his declinata in discrimina ierunt, tertia, ut ea inter se ratione coniuncta sententiam efferant, prima parte exposita de secunda incipiam hinc.'' 30. Various Connectives. Varro, L.L. Villi, 2: De huiusce multiplici natura discriminum orae sunt hae, cur et quo et quemadmodum in loquendo declinata sunt verba. ^^ " Goetz-Schoell reads ierint. ^ sunt written s. CHAPTER V Indeterminate Examples : Indirect Questions or Conditions I. Introductory: Review of Opinions concerning THE Interrogative Use or Si Opinions vary as to when the particle si came to be used to intro- duce indirect questions. Becker^ expresses himself thus: "Denique moneo si particulam, ut videtur, apud Plautum et Terentium non- dum integram interrogativae particulae vim nactam esse, sed parti- culae condicionalis naturam atque notionem ex parte retinuisse. In nonnuUis enim exemplis indicativus occurrit, in quibus si poeta particula aliqua vere interrogativa usus esset, coniunctivus flagitare- tur." Schmalz^ thinks that perhaps the first occurrence of si in the sense of num is in Propertius. Gaffiot,' similarly, denies the interrogative use of si before Propertius and Livy. Lindskog* thinks that in Plautus si never introduces an indirect question, but that in Terence there are one or two passages "ubi suspicari liceat si particulam interrogativae vi praeditam esse." On the other hand, Riemann,* Marx,' and Bennett'' hold that examples of this use occur in both Plautus and Terence.* Why this difference of opinion? It seems to me that there are two main reasons for it. In the first place, some of the scholars ' De Syntaxi Interrogatiowum ObUquarum 195. 2 Lateinische Syntax *, 519. ' Ecgui fuerit si particulae in interrogando latine usus (Paris, 1904) 30 ff.; and "Le prdtendu emploi de si interrogatif en latin," Rev. de phil., XXXII (1908) 47 £E. * De EnuntiaHs apud Plautum et Terentium Condicionalibus (Lund, 1895) 79. ^ La langue et la grammaire de Tite Live (Paris, 1885) 301; cf. also Riemann- Goelzer, Grammaire compart du grec et du latin (Paris, 1897) 409. ' "Die Beziehungen des Altlateins zum Spatlatein," Neue Jahrbiicher far das klassische Alterlum," XXIII (1909), 445. ''Syntax of Early Latin, Vol. I, The Verb (Boston, 1910), 122. ' For the conclusions reached in the present study, cf. Appendix IV. 75 76 The Indicative Indirect Question in Latin mentioned have preconceived ideas from which others are free. Becker's remark, cited above, "In nonnullis enim exemplis indica- tivus occurrit . . .," shows the ground of his opinion. Gaffiot, similarly, maintains the thesis that the indicative mood was not used in indirect questions until Augustan times.' In the second place, the terms "condition" and "indirect question" seem to be used in different ways. Thus Bennett, I.e., applies the term "indirect question" to a type of examples which other scholars, while doubtless understanding them in the same way as Bennett understands them, might classify as loosely attached conditions. Examples are Plaut. Men. 1049, Nunc ibo intro ad hanc meretricem, quamquam suscenset mihi,/Si possum exorare ut pallam reddat; Cist. 652, Cure. 701, Poen. 1064, Trin. 921, AuL 620.i» If there were generally accepted definitions of the terms "con- dition" and "indirect question," the second ground of difference would be removed. However, it would be extremely difficult, in the case of these constructions, to frame definitions which should be generally accepted, or indeed which should seem wholly satisfactory even to the person framing them. II. Clauses whose Interpretation is Indifferent 1. Introductory. In the absence of a clear statement of the difference between the indirect question and the condition, we are groping in the dark when we try to distinguish particular 5»-clauses as instances of the one or of the other construction. It will be well, then, to be extremely cautious as we proceed. It is particularly futile to attempt to label the sj-clauses which are loosely attached. Cf. Cic. De Or. II 70, 283, Vide, Scaure, mortuus rapitur, si potes esse possessor; Virg. Aen. I 578, Equidem ° Cf . "(Quis) quid relatif," 59: "Dans Plaute, comme dans Ciceron, partout oCl la suhoidonn^eestincontestablement une interrogative indirecte, elleestconstruiteavec le subjonctif." '" Further, Bennett, I.e., classes as indirect questions these two clauses which seem to me conditions: Amph. 773, Enim vero Ulud praeter alia mira miror maxima, Si haec habet pateram illam (I wonder at that more than at the other marvelous things, if she has that bowl; cf. Ter. Heaut. 525, Minimumque miror Clinia hanc si deperit), and Ter. Adel. 154, Nisi quidquid est, Volo scire atque hominem convenire, si apud forumst. Clauses whose Interpretation is Indifferent 77 per litora certos Dimittam et Libyae lustrare extrema iubebo, Si quibus eiectus silvis aut urbibus errat; Plaut. Poen. 1064, Die mihi, Ecquid meministi tuom parentum nomina, Patris atque matris? Ag. Memini. Ha. Memoradum mihi, si novi forte aut si sunt cognati mihi; Trin. 959, Enim vero ego nunc sycophantae huic sycophantari uolo, Si hunc possum illo mille nummum Philippum circumducere; Men. 1049, Nunc ibo intro ad banc meretricem, quamquam suscenset mihi. Si possum exorare ut pallam reddat quam referam domum. These passages are cited by Gaffiot, Rev. de phil., XXXII, pp. 33 £f. In all of them si may be rendered "in case," "on the chance that." Sometimes, as, for instance, in the last example, such a loosely attached 5«-clause follows a verb which does not ordinarily introduce an indirect question. I should not class such an example as an indirect question, though I should not deny that it may be one. Neither should I class the si-clauses in the other passages just cited, as indirect questions. I should rather consider that whenever a 5i-clause which occurs after a verb that may introduce an indirect question, can be paralleled by a 5»-clause which occurs after a verb that may not introduce an indirect question, the former 5J-clause may conceivably be something else than an indirect question. The examples just cited, however they may be labeled, are in any case loosely attached clauses. If they are indirect questions, they are not indirect questions of the usual sort. No examples of this kind are included in the list that is given below. There are other examples which may be regarded either as loosely attached clauses, or as indirect questions which are closely attached to the introductory verb (i.e., indirect questions of the usual kind). An instance is Cic. Att. IX 7C1, cited p. 78. Temptemus . . . si possumus may mean either, "Let us make an attempt in case we can" — the 5«-clause being loosely attached to the main verb, alike whether it is labeled as a condition or as an indirect question — or, "Let us try whether we can" — the 5f-clause being an indirect question closely attached to the main verb. There are still other examples which may be either closely attached indirect questions or closely attached conditions. Cf. Virg. Aen. I 322, cited below. Instances of both these kinds are included in the following list and are cited indiscriminately. 78 The Indicative Indirect Question in Latin 2. Examples of Clauses whose Interpretation is Indifferent. II. Inform Ter, Phorm. 675: 'Quantum potest me certiorem' inquit 'face, Si illam dant, hanc ut mittam, ne incertus siem, Nam illi mihi dotem iam constituerunt dare.' This may mean either, "Let me know whether . . .," or "Let me know it if . . ." For the following example there are similar alternatives. Virg. Aen. I 322: Monstrate mearum Vidistis si quam hie errantem forte sororum. III. Find Out Caesar in Cic, Att. IX 7 C, 1: Temptemus hoc modo si possumus omnium voluntates recuperare et diuturna victoria uti.^^ This passage may mean either, "Let us try whether . . .," or, "Let us make an attempt, in case. . . ." Cf. p. 77. IV. Know Plaut. Men. 142: lam sciam si quid titubatumst, ubi reliquias videro. This passage may mean either, "I shall know it, if there's been a mis-step . . .", or "I shall know whether there's been a mis- step. . . ." The following example is similar. Plaut Merc. 156: Quin iam prius quam sum elocutus, scis si mentiri uolo.^'^ Virg. Aen. II 739: Namque avia cursu Dum sequor, et nota excedo regione viarum, Heu! misero coniunx f atone erepta Creusa Substitit, erravitne via, seu lassa resedit Incertum, nee post oculis est reddita nostris.^* " possimus Mue. and other edd. : possumus codd., Wesenb., Boot, Tyrrell. ^scis si Camerarius: scisset {scisse P™) codd. ^* Strictly, the alternative interpretations of the i6M-clause are not indirect question and condition, but indirect question and direct question. Cf . pp. 8 fi. The example is given here with the it-clauses, because of the relationship between seu and si. Clauses Naturally Interpreted as Indirect Questions 79 VI. See Plaut. Cas. 591: Viso hue, amator si a foro rediit domum. For visere used absolutely, cf. Most. 793, Vise, specta tuo usque arbitratu. For its use with an indirect question, cf. Stich. 328, Ego quid me velles visebam; M.G. 708. Cic. Verr. Act. II, III 77, 180: Vide, quaere, circumspice, si quis est forte ex ea provincia . . . qui te nolit periisse. III. Clauses Which, ie an Indicative Indirect Question Exists, are More Naturally Interpreted as Indirect Questions than as Conditions*' 1. Indications that Clauses are Indirect Questions rather than Conditions. There are a number of ii-clauses which might be inter- preted as conditions but are more naturally understood as indirect questions. No more definite reason can generally be given for the preference of the latter interpretation than that this interpretation harmonizes better with the context. In one or two instances in later Latin (cited in Part II) the inter- pretation of the 5j-clause as an indirect question is favored by the occurrence, near to it, and in a corresponding use, of an indirect question or equivalent construction. Cf. Aug. Civ. Dei XVII, 20, Videamus ergo si sermones illius veri sunt et temptemus quae even- tura sunt illi et sciemus quae erunt novissima illius. Again, the verb may seem to demand an object, and the si-clause may be the most available object. Cf. Sen. Oed. 211, Germane nostrae coniugis, fessis opem Si quam reportas, voce properata edoce. In this, as in the preceding chapter, the .y*-clauses are, generally speaking, of two kinds. If they are not interpreted as indirect questions, some must be regarded as closely attached conditions and others as clauses that are loosely attached. An example of the former kind is Plaut. Rud. 323 (see below); one of the latter kind is Hor. Ep. I 7, 39 (p. 84). Many of the examples cited below are practically certain examples of indirect questions. Perhaps they should have been cited in " The term condition is loosely used in this chapter to designate all ^-clauses that are not indirect questions. 80 The Indicative Indirect Question in Latin Chapter VIII. Riemann-Goelzer classes most of our examples with the verb "see" as instances of interrogative si. Cf. Riemann- Goelzer, p. 409, esp. n. 5: "Ce qui prouve qu'apres videre et visere, si a bien la valeur d'une particule interrogative et n'est pas une particule conditionnelle, c'est que dans la langue correcte on trouve -ne ou num en pareil cas." However, bearing in mind the existence of loosely attached M-clauses like the one in Cic. De Or. II 70, 283 (cited p. 76), I have taken pains to include in Chapter VIII only the absolutely certain instances of interrogative si. 2. Examples of Clauses More Naturally Interpreted as Indirect Questions than as Conditions. I. Ask Plant. Rud. 329: Eadem, sacerdos Veneria haec si quid amplius scit, Si videro exquisivero; faciet me certiorem. The most natural interpretation is that sacerdos . . . si . . . scit is an indirect question depending upon exquisivero. It is less natural to regard this clause as a condition, with faciet as its conclusion, and to consider si videro exquisivero a paren- thesis. II. Inform Plant. Rud. 323: Tr. Ecquem adulescentem hue, dum hie astatis, expedite, Vidistis ire strenua facie, rubicundum, for- tem . . . Pi. Nullum istac facie ut praedicas venisse hue scimus. Tr. Ecquem , Recaluom ad Silanum senem . . . ?- Pi. Cum istiusmodi virtutibus operisque, natus qui sit, Eum quidem ad carnuficem est aequius quam ad Venerem commeare. Tr. At si vidistis dicite. Pi. Hue profecto nuUus venit. "At si vidistis dicite" is apparently a repetition in indirect form of the question, "Ecquem . . . recaluom. . . ." To Clauses Naturally Interpreted as Indirect Questions 81 explain si in this passage as a conditional particle and to inter- pret, "But if you saw him, tell me so," seems to me, as to Marx (p. 445), "eine Kunstelei." This example is not mentioned by Gaffiot. HID. Determine, Judge Plaut. As. 399: Li. Nihilo mage intus est. . . . Me. Argenti viginti minas, si adesset, accepisset. Li. Qui pro istuc? Me. Asinos vendidit Pellae mercatori Mercatu. Li. Scio. Tu id nunc refers? lam hie credo eum adfuturum. Me. Qua facie vester Saurea est? Si is est, iam scire potero. The natural meaning is, "I shall be able to determine whether he is the man"; not as Ga£6ot, Ecqui fuerit, p. 17, apparently interprets, "If he is the man, I shall be able to determine that fact." Vitruv. II 8, 19: Tunc si est firma probatur. Vitruv. VII 14, 3: De aqua . . . et qua ratione ducatur qui- busque rebus si erit salubris et idonea probetur explicabo. Morgan, On the Language of Vitruvius, p. 488, considers these two examples "conditional protases used instead of indirect questions." It seems to me necessary to regard them as indirect questions. IV. Know Plaut. Epid. 550: Novin ego te? Ph. Si ego te novi animum inducam ut tu noveris. The meaning may be either, "Do I know you? Ph. I shall give you reason to know whether I know you"; or, "Do I know you? Ph. If I know you, I shall give you reason to know (whether you know me)." The former interpretation seems to me the more natural one. VI. See Plaut. Pers. 825: Nequeo, leno, quin tibi saltem staticulum olim quem Hegea Faciebat. Vide vero, si tibi satis placet. 82 The Indicative Indirect Question in Latin Gaffiot, Ecqui fuerit, p. 14, Rev. de phil., p. 51, considers the M-clause the equivalent of "si quidem, si certe satis placet." This is possible; but it seems to me far more natural to interpret, "Just see whether it pleases you." Plant. Trin. 748: Vide si hoc utibile magis atque in rem deputas: Ipsum adeam Lesbonicum, edoceam ut res se habet." On the reading of A, Gaf&ot's interpretation — Ecqui fuerit, p. 13 — of the 5J-clause as a condition, with its conclusion in "ipsum adeam . . .," may be possible, though there is no parallel, so far as I know, for such a use of the volitive subjunc- tive in the conclusion of a condition. '^ On the reading of P, Gafhot's interpretation is still more forced: "if you think this advisable, that I go to Lesbonicus, I will inform him of the facts." From the punctuations of Goetz-Schoell, Leo, and Lind- say, it seems that these scholars feel the clause to be an indirect question. Plant. Trin. 763: Sed vide consilium si placet. Gaffiot, Ecqui fuerit, p. 14, Rev. de phil., XXXII, p. 52, would take the 5J-clause as conditional, and equivalent to sHl te plait: "Margaronides interrompt les r6flexions de Callicles, qui, un premier projet rejete, est la pensif, sans entrevoir d'issue a la situation. Aussi emploie-t-il la petite pr6caution polie si placet: c'est moins sec; cela donne du liant a la conversation." To my mind, the use of "s'il te plait" would be over-polite and not natural in this situation. I should interpret: "But look at my plan, whether it pleases you." It is possible, however, in this example, to reSort to the interpretation, "pour le cas ou" (cf. Gaffiot, Rev. de phil., XXXII, pp. 53 ff.). Ter. Adel. 239: Unum hoc habeo: vide si satis placet. Gaffiot, Ecqui fuerit, p. 19, wrests the following meaning out of the passage: "Unum hoc habeo, quod tibi proponam; si satis placet, vide quid eo facere veils (arrange- tc"' , fais-en ton " ipsum A: ufP. •' For the use of the volitive subjunctive in the first person singular, cf . Hale-Buck, Latin Grammar, §501, 2. Gaffiot calls the subjunctive potential, but, since he para- phrases with adiho and edocebo, he uses the word loosely. Clauses Naturally Interpreted as Indirect Questions 83 profit)." Cf. also Rev. de phil., XXXII, p. 51. Tyrrell, whose punctuation I follow, and Bennett, p. 122, give the s«-clause its natural interpretation, as an indirect question. Ter. Eun. 545: Homo ipse nusquamst neque scio quid dicam aut quid coniectem. Nunc mi hoc negoti ceteri dedere ut ilium quaeram. Idque adeo visam si domist. In view of the context — the speaker is looking for a man who is nowhere to be found — , it is far more natural to interpret, "And so I shall see whether he is at home," than, with Gafl&ot, Ecqui fuerit, p. 21, "And so I shall see him, if he is at home." Ter. Eun. 838: Py. . . . Vide amabo, si non, quom aspicias, OS impudens Videtur! Th. Non est? Py. Tum quae eius confidentiast ! Gaffipt, Ecqui fuerit, p. 20, would consider the s«-clause a true condition, with its conclusion in tum quae eius confidentiast. The examples which he cites of the use of tum in conclusions — Plant. M. G. 1365, Si id facies tum demum scibis, and others — are not parallel; for in them the conclusion follows as a result of the condition, while here the condition would be adversative. Even apart from the tum, the meaning which Gaffiot extracts from the passage is strained. Ter. H. T. 170: Tempust mopere me hunc vicinum Phaniam Ad cenam ut veniat: ibo visam si domist. In this context the interpretation of Gaffiot, Ecqui fuerit, p. 20, "Visam eum, saltern si domist," is less unnatural than in the preceding example. Here also, however, to me as to Tyr- rell — whose punctuation I adopt — the natural interpretation seems to be: "I shall see whether he is at home." Ter. Phorm. 553: Vide si quid opis potes adferre Jiuic. Gaffiot, Ecqui fuerit, p. 19, Rev. de phil., XXXII, p. 54, places a comma before si and interprets: "Consider, in case. . . ." The natural meaning of the passage, it seems to me, is that which it has on Tyrrell's punctuation: "See whether. . . ." 84 The Indicative Indirect Question in Latin Hor. Ep. I 7, 39: Inspice si possum donata reponere laetus. Gaffiot, Ecqui fuerit, p. 27, Rev. de phil., XXXII, p. 54, con- siders the si equivalent to 5* quidem: " 'Pais attention a tout cela, dans I'hypothese que je suis capable de rendre d'un coeur allegre les presents regus; c'est-a-dire, 'fais attention, car. . . .' " Gaffiot's interpretation, while it cannot be proved wrong, seems to me absurd. The natural understanding of the passage is that "inspice si" means "look whether," and that "si pos- sum . . ." is — as Plessis, ad loc. regards it — an indirect question. CHAPTER VI Indeterminate Examples: Miscellaneous There are a few possible instances of the indicative indirect ques- tion which do not fall into any of the preceding parts of this study. They are cited here. III. Find Out 6. Qua Causa, etc. Cic Att. Ill 23, 4: Ut Ninnium aut ceteros fugerit investiges velim et quis attulerit, et quare octo tribuni pi. ad senatum de me referre non dubitarint . . . eidem in abrogando tarn cauti fuerunt, ut id metuerent, soluti cum essent, quod ne iis quidem, qui lege tenentur, est curandum.'^ , It may be that, when Cicero vrrote fuerunt, the clause in which it occurs no longer felt to him like an indirect question. IV. Know 19. Quam. Plaut. Poen. 441 : Scin quam videtur? Credin quod ego fabuler? Thus Leo and Lindsay. Lindsay n. cr.: "Vel sic distingue: scin quam? videtur — ." Goetz-Schoell has the latter punctua- tion. 21. Unde. Plaut. Amph. 424: So. Signi die quid est. Me. Cum quadrigis Sol exoriens. Qpid me captas, carnufex? So. Argumentis vicit, aliud nomen quaerundum est mihi. Nescio unde haec hie spectavit. The meaning may be either, "I do not know from what place he saw these things" — indirect question — , or "From some place or other he saw these things." ^fuerunt oifuerant codd.: fuerint Bosius, edd. 85 86 The Indicative Indirect Question in Latin VI. See 23. Ut. Plaut. As. 149: At scelesta viden ut ne id quidem me dignum esse existumat Quern adeat, quern conloquatur quoique irato supplicet? It is conceivable that viden ut is parenthetical, and that the sentence is declarative, not interrogative. However, this is not the natural interpretation. Plaut. True. 463: Vosmet iam videtis ut ornata incedo: Puerperio ego nunc med esse aegram adsimulo. Becker, p. 246, would place, with Studemund, a colon after videtis and a comma after incedo. Becker compares Pseud. 979. CHAPTER VII The Proofs that Clauses are Indirect Questions I. The Discrimination or the Indirect Question erom All Other Kinds oe Clauses There are a number of indicative clauses which cannot possibly be interpreted as anything else than indirect questions. Any other interpretation of these clauses would yield no sense, or would yield a sense which would be impossible in the given context, or would involve the violation of some rule of grammar. The specific criteria which have been found to be of assistance in distinguishing indirect questions from other constructions are enumerated below. There is one phenomenon which distinguishes an indirect question from all other kinds of clauses, the phenomenon of illogical prolepsis.^ By illogical prolepsis is meant the use of the logical subject of a ques- tion as the grammatical, but illogical, object^ of the introductory verb. Cf. Plaut. Pseud. 1184 Chlamydem banc commemora quanti conductast. In this sentence the clause introduced by quanti must be an indirect question, for only if this clause is so interpreted will the introductory clause yield sense. By itself, "Chlamydem banc commemora" does not mean anything. II. The Discrimination or the Indirect from the Direct Question (1) The meaning of the interrogative clause itself may preclude the possibility that it is a direct question. Cf. Cic. Att. VII 26, 3, Quid ' Becker, op. cit., 165 ff., has observed that this kind of prolepsis is a criterion for distinguishing indirect from direct questions. Becker designates this kind of prolepsis by the term "artificial" {"artifkiosa"). This term is objected to by Lindskog, Quaestiones, p. 75, on the ground that the phenomenon must have originated quite naturally. Lindskog considers the examples of illogical prolepsis survivals from a time when the accusative case was used more freely than in the Latin which has come down to us. Cf. his "Zur Erklaxung der Accusativ mit Infinitiv Construction im Latein," Eranos I (1896) 132. In any case, whether these examples are survivals or products of analogy, the term "artificial" is misleading. I have, accordingly, used the word "illogical" instead. " Or, occasionally, subject. Cf. Paulus Ed. 23, 2, 10, Si fuerit ignotus ubi degit. 87 88 The Indicative Indirect Question in Latin habebo certi faciam ut scias. It would be impossible to interpret: "Of what shall I be certain? I shall inform you of it." (2) The meaning of the accompanying verb or clause may show that a following question is indirect. Thus in Plaut., M. G. 36, Scio iam quid vis dicere, it is impossible for the speaker, after he has said that he knows, to proceed to a direct question, which would be reasonable only if he did not know. For an example with a different verb, cf. Merc. 783, Dicam id quid est. Dicam cannot introduce a direct question (unless it be a directly quoted one).' Sometimes, while the introductory verb in itself is compatible with a direct question, its modifiers necessitate the interpretation of the question as indirect. Cf. Cic, Att. VII 12, 1, Sed quaeso ut scribas quam saepissime, non modo si quid scies aut audieris, sed etiam si quid suspicabere, maximeque quid nobis faciendum putas. Quaeso ut scribas might be followed by a direct question, but not quaeso ut scribas quam saepissime, . . . maximeque. (3) The logic of the larger passage of which the clause under discussion is a part, may make it impossible to interpret the clause under consideration as a direct question. Cf. Sen., Nat. Quaest. Ill 20, 2: Hae causae saporem dant .aquis varium . . . Interest, utrum loca sulphure an nitro an bitumine plena transierunt; hac ratione corruptae cum vitae periculo bibuntur.* If it were not for the logic of the larger passage to which the utrum . . . an clause belongs, it would be possible (though still not natural) to interpret: "It makes a difference. Have the waters passed through regions rich in sulphur or in natron or in bitumen?" It must be borne in mind, as we proceed, that indirect questions are to be discriminated not only from direct, unquoted questions, but also from directly quoted questions. Directly quoted questions occur with verbs which express or imply an idea of saying or of ask- ing. Cf. " 'What are you doing?' he says"; " 'What are you doing?' he asks." Cf. Plaut. Amph. 912: 'Cur dixisti?' inquies. Now, there is no possibility of confusion between the directly quoted question in " 'Quid facis?' dicit," and the indirect question in "Quid facis dicit"; for a verb of saying with an indirect question yields a totally ' See the third paragraph below. * This example, from Part 11, is cited here because there is no example in Part I to which the principle so well applies. Discrimination from the Direct Question 89 different meaning from a verb of saying with a directly quoted ques- tion. The context enables one to discriminate with certainty be- tween the two expressions. After verbs of asking the case is dif- ferent. "'Quid fads?' rogat," and "Quid fads rogat" have the same general meaning. Wherever an indirect question is used with a verb of asking,* it is conceivable that a directly quoted question should have been used instead. In the case of an indicative ques- tion which is used with a verb that expresses or impUes an idea of asking, it is, accordingly, impossible to determine with absolute certainty whether the question is indirect or is directly quoted.' There is an exception to this statement, and that is in the case of double questions: (4) If, in a double question of fact, the indicative occurs in one half and the subjunctive in the other, the indicative verb must, like the subjunctive verb, be in an indirect question. Cf. Asconius in Cornelianam 54 (cited p. 144). It would be impos- sible to regard one half of the question as indirectly quoted and the other half as directly quoted. The subjunctive shows that the one half is indirectly quoted. Therefore, the other half must be indi- rectly quoted also. After verbs other than those which express or imply inquiry, if we know that a question is dependent, we know also that it is indi- rect.' Except after a verb of inquiry, a proof that a question is dependent is a proof that it is indirect. The following circum- stances, then, which show that questions are dependent, show also (with the reservation mentioned) that they are indirect. ' In such a case, an indirect question is an indirect quotation (cf. p. 4, n. 4). For the difference between direct and indirect quotation, cf. J. J. Schlicher, "The Moods of Indirect Quotation," A J P XXVI (1905), 69: "Direct quotation is the repetition of the form in which the original speaker expressed certain ideas, while indirect quotation is the expression by the hearer of these ideas as they lie in his own mind." ° For the distinguishing of indirect questions by a shift of tense or person, see p. 91. ' After a verb of saying, to be sure, a dependent, directly quoted question may occur; but, as was explained above, such a question would have a totally different meaning from an indirect question, and — unless it were apart from its context — could not be confused with one. It may, accordingly, be left out of account in this dis- cussion. 90 The Indicative Indirect Question in Latin (5) A question is dependent if it breaks into the midst of the sentence that introduces it.* Cf. Plant. Pseud. 696a: Id tu modo me quid vis facere, fac sciam, where the question Me quid vis facere breaks into the introductory clause, Id tu modo fac sciam.^ It is not, however, always possible to determine whether a question does break into the introductory sentence. In Pers. 640, Sed tamen, virgo, quae patriast tua age mihi actutum expedi, sed tamen may con- ceivably be felt with the question, and not with expedi. And in a sentence like Capt. 964, Tandem istaec aufer, die quid fers, ut feras hinc quod petis, or Men. 779, Loquere uter meruistis culpam, paucis, non longos logos, the part that follows the question may be an afterthought. In one instance the sentence that is interrupted by the clause under consideration consists of an introductory verb and a subjunc- tive indirect question. This instance is Ad Heren. I 6, 10, Quid alii soleant, quid nos facturi sumus, breviter exponemus. In such an example, it is not the parallel use of a subjunctive indirect question that proves the indicative question to be dependent, but rather the fact that the indicative question interrupts the sentence.'" (6) If two or more questions of detaiP"^ are connected by correla- tive conjunctions — et . . . et, aut . . . aut, neque . . . neque — they must be dependent. Cf. Plant., Cist. 57, Eloquere utrumque nobis, Et quid tibist et quid velis nostram operam, ut nos sciamus. (7) A clause is dependent if it is connected, by a conjunction, to a noun which is in a grammatical relation with some verb. Cf. Tibull. II 4, 18, Nee refero solisque vias et qualis, ubi orbem Com- plevit, versis Luna recurrit equis. * It is, to be sure, conceivable that a sentence should be broken into by a paren- thetical, independent question. However, in none of the examples to which criterion 5 applies, would it be possible to interpret the question under consideration as paren- thetical. Lindskog, Quaestiones, 95, has observed a connection between the position of clauses and their dependence or independence: "Id solum hie velim dicere, quo magis ita excoleretur lingua latina, ut hypotactica structura pro paratactica uteretur, eo magis earn collocationem, qua insereretur secundaria in primariam, amatam esse." '" If the order of the clauses were different, and the conjunction et were used, it might be possible — though it would not be natural — to interpret: Quid nos facturi sumus? Breviter exponemus, et (sc. exponemus) quid alii soleant. " It is conceivable that the connectives mentioned should join "yes and no ques- tions." A combination Uke Et iu ad eum isti et is ad te venit? is possible. Discrimination from the Exclamation 91 It may be thought that for questions introduced by num, there is still another criterion. When it introduces a direct question, num suggests a negative answer, but when it introduces an indirect ques- tion, it is non-committal.^^ Any indicative »Mw-questions for which a negative answer is clearly not suggested, would, accordingly, neces- sarily be indirect. However, I know of only one »Mw-question in which a negative implication seems impossible: Prop. Ill 5, 43 (cited pp. 17 ff.); and even apart from this circumstance, the question can hardly be direct. In some languages, as EngUsh, German, and French, the position of the verb in the question may indicate whether the question is direct or indirect. In Latin, however, this seems not to be the case." It is sometimes said that a shift in person or tense may show that a clause is dependent. Eduard Hermann, in "Gab es im Indo- germanischen Nebensatze," K. Z. Z., XXXIII (1895), p. 486, opposes this view. In "A laesst dir (C) sagen, du (C) sollst zu ihm kommen," the second clause may be independent. If it is dependent, we feel it to be so not because of a shift in person, but "wegen der kurzen Pause zwischen beiden Satzen, und wegen der monotonen Aus- sprache des zweiten." For the use that has been made in this study, of shifts in tense or person, see p. 25, criterion 4. Frank, Attraction of Mood in Early Latin, p. 28, remarks that in . the direct question in Plautus, the feminine substantive form of the interrogative pronoun is quis, while in the indirect question it is quae. This observation gives no help in distinguishing indirect from direct indicative questions; for none of our apparent instances of the indicative indirect question from Plautus contains a feminine sub- stantive interrogative pronoun.^^ III. The Discrimination of the Indirect Question FROM the Exclamation The proofs that clauses are not exclamations are practically the same as the proofs that clauses are not direct questions. Cf . pp. 87 fi. " Cf. Hale-Buck §§231, 1 d and 537 d 2. '' My observations confirm the conclusion of Morris, Principles and Methods in Latin Syntax (New York, 1901) 125, that in Latin "there appears to be no tendency to indicate subordination by the position of the verb." " The only instance that I have noted of an indicative indirect question intro- duced by a feminine substantive pronoun is Buecheler 73, 1 (cited p. 99). Here the form of the pronoun is quae. 92 The Indicative Indirect Question in Latin (1) The content of the clause under debate may preclude exclama- tory feeling. So in Varro, R. R. II 5, 9 (cited p. 104) the meaning of the quibus-c\3i.nse in its context is such that the clause would not be uttered in an exclamatory tone. It would be impossible to punc- tuate: Et praeterea quibus regionibus nati sunt! Refert. (2) The meaning of the accompanying word or clause may demand that the clause under consideration be regarded as an indirect ques- tion. Cf. Publilius Syrus, Ribbeck II, p. 339, 348, Magis valet qui nescit quid valet calamitas. (3) The logic of the larger passage which contains the clause may demand that the clause be understood as an indirect question. Cf. Plant Pseud. 263a, Nosce saltern hunc quis est. Ba. lam diu scio Qui fuit: nunc qui is est ipsus sciat. In this example, if the sentence, "lam diu scio Qui fuit" were considered by itself, it might be possible — though it would not be natural — to regard the qui- clause as an exclamation: "I have long known it — what kind of man he has been! The larger context, however, makes it necessary to regard the clause as an indirect question. (4) To prove that a clause is not an exclamation, it is sufficient to show that it is dependent^^ and is not a quotation. For proofs that clauses are dependent, cf. pp. 89 fi., criteria 4-6. A quoted exclamation may occur only after a verb that expresses or implies an idea of saying. A possible — but not probable — instance of a directly quoted exclamation is Ter., Hec. 472, cited p. 33. IV. The Discrimination or the Indirect Question FROM the Relative Clause A dependent indicative clause that is introduced by an indeter- minate form of the relative-interrogative pronoun,^' or by any relative-interrogative connective, can be certainly recognized as an indirect question, by the following circumstances. (1) Disagreement in number or gender between the introductory word of the dependent clause and the word in the main clause — " Outside of quotation, there is no such thing as a dependent exclamation that is distinct from a dependent question. Cf. p. 29, n. 1. " All forms are indeterminate except quis and quid; and sometimes these are indeterminate. Cf. Appendix III. Discrimination from the Relative Clause 93 expressed or understood — with which the introductory word, if it were relative, would agree. Cf. Cic, Har. Resp. 17, 37, Obscure dicitur, quae sacra polluta sunt (sint some codd., cf. p. 97)." (2) The presence in the main clause of a word with which the dependent clause as a whole must be in apposition. Cf. Plant. Bacch. 720, Ch. Quid tu loquere? Mn. Hoc, ut futuri sumus. If the dependent clause were relative, the passage would mean, "(Lo- quor) ita ut facturi sumus" (cf. n. 18, below). In that case hoc could not be used. (3) The circumstance that the dependent clause, if interpreted as relative, would yield either no sense or a different sense from that demanded by the context. Cf. Cato, Agr. Cult. VI 4, Vineam quo in agro conseri oportet, sic observato (The meaning is certainly not: "Observe the field in which."); Ad Keren. IV 9, 13, Quibuscum hel- ium gerimus, indices, videtis: cum sociis, qui pro vobis pugnare . . . soUti sunt; Prop. II 30, 29, Illic aspicies scopulis haerere Sorores Et canere antiqui dulcia furta lovis, Ut Semela est combustus, ut est deperditus lo, Denique ut ad Troiae tecta volarit avis. Ut can- not be relative, for the meaning "in such a way as" would not be possible. '* " The objection may be raised, that if "Nota enim mihi sunt, quae in me post Caesaris mortem contulerint," is a contamination of "Notum est, quae contulerint" and "Nota sunt, quae contulerunt" (cf. my "The Confusion of the Indirect Question and the Relative Clause in Latin," Classical Philology XIII [1918] 66), "Obscure dici- tur, quae sacra polluta sunt" may be a contamination of "Obscure dicitur quae sacra polluta sint," and "Obscure dicuntur quae sacra polluta sunt." However, it would, as it seems to me, be unjustifiable to resort to this complicated interpretation when a simple one is available. It would seem unjustifiable, too, to interpret the passage under discussion iil the light of the rare, and somewhat irregular, used of the relative clause found in Trin. 1123, Nunc is propere conveniendust ut quae cum eius filio Egi ei rei fundus pater sit potior. (That tlie matters which I have transacted with his son, that this arrange- ment may be officially confirmed by the father.) To interpret "Obscure dicitur quae sacra polluta sunt," "The sacred rites which have been violated, is this matter not clearly stated?" would be absurd. " If ut is relative, it must mean "in such a way as;" "in the way in which." Gaffiot, to be sure, is of a different opinion. He remarks, "(Quis) quid Relatif," Rev. de phil., XXXIV (1910) 60, that when Latin says, Video ut res gesta est, "il envisage at comme relatif; et nous le rendrons analytiquement par 'la manifere dont.' " This view is opposed by the fact that, so far as I can find, in all the certain examples of the lelative ut, except in stereotyped combinations like praeut, the meaning is not la 94 The Indicative Indirect Question in Latin The last criterion must be used with caution. As language is often not logical, the value of this criterion depends upon the range of observation of the person who uses it. How cautious one must be, may be seen in the case of examples like Most. 459, Non potest Dici quam indignum facinus fecisti et malum. It might well seem that a tain in this passage would make no sense, and that, therefore, the gwaw-clause must be an indirect question. But observe the relative clause in Ter.,Hec. 417, Non hercle verbis, Parmeno, dici potest Tantum quam reapse navigare incommodumst.^' This clause resembles the one in Most., and it is, therefore, possible — though not probable — that in Most, a tarn or tantum is to be supplied in thought.^" V. The Discrimination or the Indirect Question FROM THE CONDITION^l There are a few examples of 5i-clauses which cannot possibly be anything else than indirect questions. These are known to be indirect questions for the following reasons. (1) The occurrence, parallel to the sirclause, of an indirect ques- tion or equivalent construction, to which the s4-clause is attached by means of a conjunction. Cf. Prop. II 34, 53: Harum nulla solet rationem quaerere mundi, Nee cur fraternis Luna laboret equis. Nee si post Stygias aliquid rest arbiter undas. Nee si consulto fulmina missa tonent. Cf. also, in later Latin, Greg. Passio Sept. Dorm. 7, 'Enarra, quaesumus, nobis, frater, quae hac nocte locutus est impera- tor, aut si inquisiti sumus, ut sciamus.' Alike whether the quae- clause is interrogative or relative, the si-clause used parallel to it must be an indirect question. (2) The occurrence, following the si-clause, of an an or necne with which the verb of the 5i-clause is understood. Cf. Lygdamus maniere dont, but de la manUre dont, eo modo quo. There are no examples of relative lit with a word like modum as its antecedent. If ut is ever relative in the combination videre ut, the meaning must be, "I see as (i.e., in the way in which)." Cf. Ego video ut tu vides (an imaginary example). " Quantum for quam A. reapse TjTrell: re ipsa codd. 2« Cf. p. 55, n. 27. ^ The term condition is here loosely used, to designate all ii-clauses that are not indirect questions. Discrimination from the Condition 95 in TibuU. Ill, 1, 19, Ilia mihi referet, si nostri mutua curast, An minor, an toto pectore deciderim. (3) The circumstance that the ii-clause is in apposition with the subject or object of a verb of inquiry, or is used to explain a noun like quaestio. Cf. Livy XXXIV 3, 5, Nulla lex satis commoda omnibus est; id modo quaeritur, si maiori parti et in summam pro- dest. Cf. ialso Tert., Adv. Hermog. 27, cited p. 144. (4) The fact that the main verb of the sentence cannot be used absolutely and needs the 5i-clause, — which in that case must be an indirect question — as its object. Cf. Greg. Mirac. B. Andr. Apost. 28, "Non gustabo donee cognoscam, si miserebitur Deus huic homini, et si sit reputandus inter salvatos."^^ I can find, in the Thesaurus, no example of cognosco used absolutely, except in its technical, legal meaning. (5) The circumstance that no kind of j«-clause except an indirect question will yield the meaning demanded by the context. Cf. Ter. Hec. 321, Uxorem Philumenam Pavitare nescio quid dixerunt: id si forte est nescio. The meaning cannot be, "If it is that, I do not know it." Neither is Gaffiot's interpretation (cited p. 103) possible. ^ This example, from Part II, is cited, because it illustrates the principle better than any of our earlier examples. CHAPTER VIII Clauses Which, if the Sole or Best Manuscript Tradition is Followed, are Certainly Indicative Indirect Questions^ I. Ask 28. Si. The only absolutely certain examples of indicative indirect questions used after verbs meaning "ask" are 5J-clauses. After such verbs questions introduced by connectives other than si are, in the nature of the case, conceivably directly quoted ques- tions. Cf. p. 89.2* Prop. II 34, 53: Harum nulla solet rationem quaerere mundi, Nee cur fraternis Luna laboret equis. Nee si post Stygias aliquid rest arbiter undas, Nee si consulto fulmina missa tonent.^^ This M-clause is conceded by GaflSot, Ecqui fuerit, p. 36, to be an indirect question. However, he prefers the reading restaverit. Livy XXXIV 3, 5: Nulla lex satis commoda omnibus est; id modo quaeritur, si maiori parti et in summam prodest. In this example, id must be filled out, and there is nothing to fill it out except the «-clause. If the 5i-clause supplements id,, it can be interpreted in no other way than as an indirect question. This example was overlooked by Gaffiot, Ecqui fuerit, pp. 30 fi. *' It has been my endeavor to cite here only the absolutely certain examples. There are some clauses in earlier chapters which are practically certain instances of the indicative indirect question. Cf. p. 159. " There are two exceptions to this rule (cf. p. 89, criterion 4 and p. 91, top), but I have found no instances in republican or Augustan Latin, which would come under these exceptions. i» rest arbiter undas Munro (aliquis sedet arbiter Jacob) : restabit erumpnas L P (erumpnas om. N): restaverit undas D V: restabitur Phillimore: restabimus Wassen- bergh: restabit arenas a scholar cited by Burmann. 96 Clauses Which are Certainly Indirect Questions 97 11. Inform 1. Quid. Plaut. Merc. 783 :^ Fortasse te ilium mirari coquom Quod venit atque haec attulit. Dicam id quid est." Ad Heren. 16, 10: Quid alii soleant, quid nos facturi sumus, breviter exponemus.^'^ Cic. Att. VII 12, r. Sed quaeso ut scribas quam saepissime, non modo si quid scies aut audieris, sed etiam si quid suspicabere, maximeque quid nobis facien- dum aut non faciendum putas}^ Gaffiot classes this example and the two examples following, as relative clauses. Cf. Appendix, pp. 191 ff. Cic. Att. VII 26, 3: Quid habebo certi faciam ut scias.^' Att. XI 19, 1: Velim, quid erit, qualecumque erit, scribas.'" 4. Quae, etc. Har. Resp. 17, 37: Obscure dicitur, quae sacra poUuta sunt?" 8. Qualis, etc. Varro Reliquorum De Gramma tica Libiorum Frag- menta, p. 225 (Keil G. L. VII 29, 3) : Qui transscribunt tabulas, non describunt, sed exscribunt; qui quales sunt scribunt, ii describunt.'^ Tibull. II 4, 18: Ite procul, Musae, si non prodestis amanti: Non ego vos, ut sint bella canenda, colo, Nee refero solisque vias et qualis, ubi orbem Complevit, versis Luna recurrit equis." " For this example, cf. Appendix, p. 199. " id quid A: quid P. Becker (p. 245) suggests id quod est. "a Thus Marx, with the tmitUi and d: simus b, 1. "Thus Gaffiot {Pour le vrai latin, p. 70) with the codd.: putes Lambinus, edd. "Thus Gaffiot (p. 68) with the codd.: quod edd. "> Thus Gaffiot (p. 70), with Mediceus: quidquid erit or si quid erit, edd. " sunt G E: sint other codd., edd. " sunt codd.: sint Keil, edd. " recurrat dett. 98 The Indicative Indirect Question in Latin This is the only example in Tibullus of the indicative indirect question. Cf. Wolff, De Enuntiatis Interrogativis, p. 39. 10. Quantus, etc. Plant. Pseud. 1184: Ha. Quid maneam? Ba. Chlamydem hanc commemora quanti conductast. Ha. Quid est? Becker (p. 169) emends. This is an example of illogical prolepsis. For illogical prolepsis cf. p. 87, n. 1. Cic. Off. II 7, 23: Nee vero huius tyranni solum interitus declarat, quantum odium hominum valet ad pestem, sed reliquorum similes exitus tyran- norum.'* Livy XXII, 36, 1 -^^ Quantae autem copiae peditum equitumque additae sunt, adeo et numero et genere copiarum variant auctores, ut vix quicquam satis certum adfirmare ausus sim.'* 19. Quam. Caecil. Statius, Rib- beck, Up. 42,43: Nam quam duriter Vos educavit atque asperiter, non negat. Becker, p. 245, would change, with Bothe, to educarit. 23. Ut. Plant. Bacch. 720: Ch. Coctumst prandium? Vos duo eritis atque amica tua erit tecum tertia? Mn. Sicut dicis. Ck. Pistoclero nulla amicast? Mn. Immo adest. Alteram ille amat sororem, ego alteram, ambas Bacchides. Ck. Quid tu loquere? Mn. Hoc, ut futuri sumus. ^ valeat Mue. and other edd. Baiter-Halm n. cr.: "valeat c valet B H a b o." 5* Livy XXXI 7, 8, contains an additional example of the indicative indirect question, if the manuscript reading is accepted. However, it seems to me that modem editors are probably right in inserting aeguabitis and regarding the quantum-cLa.\ise as an exclamation: Ne aequaveritis Harmibali Philippum, ne Carthaginiensibus Mace- donas; Pyrrho certe aequabitis. "Aequabitis" dico? Quantum vel vir viro, vel gens genti praestat! " sunt PCM: sint other codd., edd. Clauses Which are Certainly Indirect Questions 99 Capt. 236: Nunc ut mihi te uolo esse autumo. The verb autumare sometimes introduces an indirect state- ment, as in Amph. 306, Quattuor vires sopori se dedisse hie autumat. Sometimes it has an object, as in Capt. 897, Si vera autumas. There are apparently no examples of autumare used absolutely, as dicere is used in Cic. Pro Milone 10, 29 (cited p. 63). Accordingly, the M^clause in this example is not to be interpreted like the Mi-clauses on pp. 63 £f ., as possibly relative but more probably interrogative. If autumare is never used absolutely, this M<-clause must be an indirect question. Plant. M. G. 1289: Mitto iam ut occidi Achilles civis passus est.'" Prop. II 30, 29 : Illic aspicies scopulis haerere Sorores Et canere antiqui dulcia furta lovis, Ut Semela est combustus, ut est deperditus lo, Denique ut ad Troiae tecta volarit avis. Prop. II 34, 36: Nam rursus licet Aetoli referas Acheloi, Fluxerit ut magno fractus amore liquor, Atque etiam ut Phrygio fallax Maeandria campo Errat et ipse suas decipit unda vias, Qualis et Adrasto fuerit vocalis Arion Tristis ad Archemori funera victor equus. These and other examples from Propertius are rightly inter- preted as indirect questions by Uhlmann, De Sexti Properti Genere Dicendi (Miinster, 1909) p. 74. 28. Si. Lygdamus in TibuU. Ill 1, 19: Ilia mihi referet, si nostri mutua curast An minor, an toto pectore deciderim. Gaffiot, Ecqui fuerit, p. 34, considers the sf-clause a condition. But what, in that case, does "an minor" mean? It seems to me impossible to regard si as anything else than interrogative. III. Find Out 4. Quae, etc. Buecheler 73, 1 (CIL IX 1527) : Hopes resiste et quae sum in monumento lege. Mommsen, in CIL, comments: "litteris antiquis." »' 1287-1289 om. A. 100 The Indicative Indirect Question in Latin 28. Si. Prop. Ill 5, 40: Turn mihi naturae libeat perdiscere mores, Sub terris sint iura deum et tormenta Gigantum, Tisiphones atro si furit angue caput, Aut Alcmaeoniae furiae aut ieiunia Phinei, Num rota, num scopuli, num sitis inter aquas, Num tribus infernum custodit faucibus antrum Cerberus, et Tityo iugera pauca novem An ficta in miseras descendit fabula gentis, Et timer baud ultra quam rogus esse potest. Gaffiot, Ecqui fuerit, p. 37, acknowledges that this is an indi- rect question. He thinks that Propertius is using a Grecism (p. 39, Le prStendu emploi, p. 47). IIIB. Observe 28. Si. Vitruv. I 73, 17: Neque animadvertunt si quid eorum fieri potest necne. Morgan, On the Language of Vitruvius 488, classes this example among "conditional protases used instead of indirect questions." However, necne clearly shows that the si-clause is an indirect question. HID. Determine, Judge 10. Quantus, etc. Cic. Acad. Prior. II 15, 46: Adversatur enim primum, quod parum defi- gunt animos et intendunt in ea quae perspicua sunt, ut, quanta luce ea circumfusa sunt possint agnoscere.'^ HIE. Consider, Reflect 10. Quantus, etc. Cic. Verr. Act II, I 8, 32: Intelligere debetis primum interesse utrum id onus vosmet ipsi reieceritis, an . . . vobis iudicandi potestas erepta sit; deinde etiam illud cogitare, quanta periculo venturi sumus ad eos " suni V A B, Orelli: sivi other codd., Davies, Halm-Baiter, Mueller. Clauses Which are Certainly Indirect Questions 101 iudices, quos propter odium nostri populus Romanus de nobis voluerit iudicare.^' IV. Know Becker has a theory that scio (pp. 256 £E.) and similarly scin (pp. 276 ff.) were sometimes used by Plautus like nescio, in close association with an interrogative pronoun, and without any influence upon the following verb. In this way he tries to account for the use of the indicative mood in some of the exam- ples cited below: e.g., Aul. 174. This theory does not seem to me to square with the facts. Indeed, Becker himself acknowl- edges (p. 258) that in questions depending upon scio Plautus appears "certas leges non observasse." Cf., against Becker's theory, Morris in A. J. P., X (1889), p. 412. 1. Quid. Plant. Aul. 174: Scio quid dictura's: hanc esse pauperem: haec pauper placet. Bacch. 78: Scio quid ago. Pi. Et pol ego scio quid metuo. Merc. 431: De. At ego — . Ch. Quin ego, inquam — . De. Ah nescis quid dicturus sum, tace."° M. G. 36: Scio iam quid vis dicere. True. 862: Scio mecastor quid vis et quid postules et quid petas.*" Ter. Adel. 996: Tibi, pater, permittimus: Plus scis quid opus factost." Publilius Syrus, Ribbeck II, p. 339, 348: Magis valet qui nescit quid valet calamitas.*^ " sumus G 12f (according to Halm-Baiter), sirims other codd., edd. "'a sim Jtal., favored by Becker, p. 219. *" uis cod., Lindsay: uelis Camerarius, Ritschl, Leo, G.-S. ''Thus S: sit D 'L: quod optis factost Fleckeisen, Becker (p. 254), Tyrrell (with comment: "Post quid potius sequeretur coniunctivus") : Quid facto opus sit Dziatzko- Kauer (ed. 1903). *■ quod F, Wolfflin, Ribbeck. 102 The Indicative Indirect Question in Latin Cic. Att. I 19, 4: Att. X 12, 4: Cat. 113, 5: Ov. Her. VII 53: Met. X 637: LivyXLI24, 5: 2. Quis. Plaut. M. G. 925: Metellus est consul sane bonus et nos admo- dum diligit; ille alter nihil ita est, ut plane quid erit nesciat.^' Servium exspecto, nee ab eo quicquam iyies. Scies quid erit.** Atque hoc etiam sunt timendi magis, quod quid cogitant me scire sentiunt, neque tamen permoventur.*^ Quid, si nescires, insana quid aequora possunt? Expertae totiens tarn male credis aquae?^* Dixerat, utque rudis primoque cupidine tacta. Quid facit ignorans amat et non sentit amor em.*' Nos autem qui nee ob quam causam nec- quemadmodum perierit Demetrius scimus, nee quid Philippus, si vixisset, facturus fuerat, ad haee, quae palam geruntur, eonsilia nostra ad- commodare oportet.** Pal. Num ille te nam novit? Acr. Numquam vidit: Qui noverit me quis ego sum?*' 4. Quae, etc. Varro L. L. V 31, 140: Plaustrum ab eo quod non ut in his quae supra dixi, sed omni parte palam est, quae in eo vehuntur, quod perlucet, ut lapides, asseres, tignum.^" " Thus Gaffiot, Pour le wai latin 71, with the two editions of the fifteenth cen- tury: quid emerit edd., following Cratander. See Gaffiot's note. Tyrrell-Purser n. cr.: "quidem (vel quid) erit (vel est vel sit) codd." ** quid M, GaflSot, op. cit. 69: quicquid W C, edd. '^ quid cogitant a o Lg. 50, 53, 57: quidguid cogitant e, Halm-Baiter: quidquid cogitent d w: quid cogitent other codd., Lambinus. " possint E s. *' Quid facit O X, Merkel, Magnus: quod facit dett. *' Thus Drakenborch: fuerit Weissenbom-Heraeus. *' sum B D: sim C, edd. "•Thus Goetz-SchoeU with codd. Spengel n. cr.: "scrib. vid. aut vehantur aut palam sunt." Clauses Which are Certainly Indirect Questions 103 Cic. Fam. XIV 5,1: lam enim me ipsum exspectas sive nos ipsos, qui quidem quam primum ad vos venire cupi- mus, etsi, in quam rem publicam venimus, intellego.^^ 5. Qui (Nominative Singular Substantive; cf. p. 56, n. 29). Plant. Pseud. 263a: Fs. Nosce saltern hunc quis est. Ba. lam diu scio Qui fuit: nunc qui is est ipsus sciat.^^ Becker, p. 310, interprets qui fuit as relative: "Mente supple: scio eum talem, qui." This seems to me impossible. 19. Quam. Caelius in Cic. Fam. VIII 15, 1 : Si scias quam sollicitus sum, tum banc meam gloriam, quae ad me nihil pertinet, derideas.^' 23. Ut. Plant. Stich. 112: Edepol, pater, Scio ut oportet esse, si sint ita ut ego aequom censeo. Cic. Att. Ill 7, 3: Quem quidem ego nee quo modo visurus nee ut dimissurus sum, scio.^* 28. Si. Ter. Hec. 321: Uxorem Philumenam Pavitare nescio quid dixerunt; id si forte est nescio. Pamphilus and his slave Parmeno are about to enter Pam- philus' house. At vs. 315 they hear a noise of people bustling about within. Pamphilus expresses his apprehension. Then Parmeno speaks the words cited above. The meaning is clearly: "I don't know whether it is perhaps that [i.e., a shivering fit of Philumena's] which causes the excitement." Gaffiot, Ecquifuerit, p. 21, rejects this simple interpretation, as given in the edition '^venianms Lambinus, Mue., other edd.: venimus codd. (according to Purser), Gutsche {De Interrogationibus Obliquis apud Ciceronem, p. 112), Gaffiot (p. 24). "jwi is est Ritschl: qui sit ipse A: qui estis ipsus P: quis Camerarius: quis est ipsus Lbidsa.y. Becker, p. 253, would read quis is sit. "sim Mue. and other add.: sum M, Boeckel ("Epistulae Selectae," [Heidelberg, 1908]). " sim Mue. and other edd. Purser n. cr. : "sim s sum codd. cett." 104 The Indicative Indirect Question in Latin of Thomas, in favor of the following impossible one: "Parmeno ... a domino . . . cui mens est vehementer soUicita, interro- gatus, incerta consulto respondet, quo melius impetum declinet: 'pavitare nescio quid dixerunt' ; statimque, qui culpa non vacare se senserit, iterando inscientiam confirmare studet: 'id,' hoc est, 'eum pauorem nescio' ; atque adeo, ut se dubitanter loqui planius etiam significet, illam sententiam interponit 'si forte est.' " . ^ VI. See 4. Quae, etc. AdHeren. IV9, 13: In mediocri figura versabitur oratio . . . sic: Quibuscum bellum gerim'us, indices, vide- tis: cum sociis, qui pro vobis pugnare . . . soliti sunt." 10. Quantus, etc. Cic. Fam. VII 4: Vides enim, quanto post una futuri sumus." 19. Quam. Cic. Fam. VI 9, 2 : Quam mihi necesse est eius salutem et f or- tunas quibuscumque rebus possim tueri, vides.'' 30. Various Connectives. Varro, L. L. X 2, 9: Quare quae et cuius modi sunt genera similitudinum ad hanc rem, perspiciendum ei qui declinationes verborum proportione sintne quaeret.^* . ^ , VII. Concern 4. Quae, etc. Varro R. R. II 5, 9: Et praeterea quibus regionibus nati sunt refert.5' The total number of the certain examples of the indicative indirect question may appear small. But cf. p. 96, note 23. " Muller n. cr. : "geramus editores, Langen Phil. 37, p. 405." Langen n. cr. "Solus Monasteriensis veram scripturam servavit, quae est geramus." Marx, (p. 37), thinks that the superior leadings which Langen quotes from the codex Monas- teriensis are "coniecturae . . . doctorum medii aevi." " simus Mue. and other edd. Purser n. cr. : "simus Vict, swmus M R." Gutsche (p. 112) quotes indicative. Skutsch, Glotta III 366, remarks that the clausula supports the indicative. " Thus Gaffiot, Pour le vrai latin 38, with the codd.: sit Lambinus, edd. " sint Spengel. " sint Keil, Goetz: sunt codd. CHAPTER IX Preliminary Conclusion There are found, in the Latin of the republican and Augustan periods, a large number of indicative clauses which are most naturally interpreted as indirect questions. Further, there occur a smaller, but still considerable, number of clauses which, if the sole or best manuscript tradition is followed, must unquestionably be interpreted as indicative indirect questions. 105 CHAPTER X The Disposal of the Indicative Indirect Question BY Emendation When an apparent example of the indicative indirect question cannot possibly be interpreted as a direct question, an exclamation, a relative clause, or anything else than an indirect question, emenda- tion is often resorted to. Cf. p. xviii, n. 7; p. xix, n. 11; p. xxi, n. 14; critical notes passim. It is true that frequently a very sUght change will dispose of an instance of the indicative indirect question. In many an example nothing more is necessary than to make such an easy substitution as that of sint for sunt. Again, the reading of quod in place of quid will often change an indirect question into a relative clause. Cf., e.g., the passages cited on pp. 101 ff. Indeed, a large number of the examples of the indicative indirect question might, without any very violent changes, be transformed into subjunctive indirect questions or into relative clauses. However, there are some passages among our certain examples of indirect questions, which, for metrical reasons, cannot possibly be emended, or at least cannot be emended without violence. Such passages are Plaut. Stich. 112 (p. 103); Pseud. 1184 (p. 98); Ter. Hec. 321 (p. 103); Prop. II 30, 29 and II 34, 36 (p. 99).i These passages alone, it seems to me, establish beyond the possibility of doubt, the existence of the indicative indirect question in Latin. Confirmatory evidence of the use of the, indicative mood in indi- rect questions is found in inscriptions. To be sure, there seem to be no instances of this use in inscriptions of the republican period. I ' Similarly, the rhythm of a prose passage may support the indicative reading. In Cic. Att. II 10 and Fam. VII 4 the indicative, as Skutsch has pointed out, yields a better clausula than the subjunctive would do. See pp. 15 and 104. For the clausulae involved, cf. Zielinski, Der constructive Rhythmus in Ciceros Reden (Leipzig, 1914) 652 (clausula V 2) and 739 f. (S 2). One should, however, be cautious about using clausulae as criteria for the text-criticism of Cicero's letters. So far as I am aware, the rhythm of these letters has not been subjected to the careful study which has been devoted to the rhythm of the orations. 106 Disposal of the Indirect Question by Emendation 107 have not found any in Volume I of the Corpus. Since, however, there are only two examples, so far as I have observed, of the sub- junctive indirect question — CIL., I, 1220 and 1479 — the absence of the indicative construction is not significant. In inscriptions of the empire, the indicative indirect question occurs not infrequently. Examples taken from various collections are cited in this paper. I have read the Monimentum Ancyranum in search of indirect ques- tions, but I have found no examples either of the indicative or of the subjunctive construction. The inscriptions of Pompeii, Stabiae, and Herculaneum in CIL., IV and X — which I singled out for read- ing because their terminus ante quern is fixed — furnish two examples of the subjunctive indirect question — CIL., IV Suppl. 4971 and 5242 — and none of the indicative. Buecheler, however, cites an example from Pompeii of an indicative indirect question (Buecheler 932, from Notizie dei Scavi; see below, p. 129). W. S. Fox, "The Johns Hopkins Tabellae Defixionum," Supplement to A. J. P., XXXIII 1 (1912), p. 42, says that he finds only one example of the indirect question in the Defixiones. He cites this on p. 17, 1. 27. A lacuna renders it uncertain whether the indicative or the subjunctive was used. In view of the occurrence of instances of the indicative indirect question, which it would be impossible to "emend," one should, it seems to me, be very cautious about "emending" examples of this construction, even when a change would be easy. One should not change quid to quod, or sunt to sinf, for the sake of disposing of an example of the indicative indirect question. To be sure, quid and quod, and sunt and sint have often been confused by scribes. Never- theless, while some instances of quid with the indicative may be due to a copyist's mistaking the sign for quod, and some instances of sunt may be due to a similar error, yet in all probability the instances in which sunt has been changed to sint, or quid to quod, in clauses of the kind which we are studying, are more numerous than the instances in which the contrary changes have been made. Because the sub- junctive was the more usual mood in indirect questions (see p. 169) a scribe would be more apt to make the mistake of copying an indica- tive as a subjunctive, than to commit the opposite error.^ Similarly, ' Cic. Rep. I 19, 31 (cited p. 12) seems to be an example of a scribe's miscopying or wilfully changing an indicative indirect question. Here the false subjunctive was 108 The Indicative Indirect Question in Latin he would be more apt to copy a quid introducing an indicative indirect question as quod, than vice versa. As regards intentional changes, it is hardly conceivable that a subjunctive in an indirect question should ever have been purposely changed to an indicative. It may be argued that at any rate in particular classes of writings the certain examples of the indicative indirect question are so very rare as to make emendation permissible. It is true that the scholar who emends the few examples of the indicative indirect question that occur in the formal prose-writings of Cicero, may conceivably be restoring what Cicero wrote.' In all probability, however, this is not the case. As was remarked above, an original indicative in. an indirect question would be more apt to be changed by a scribe to a' subjunctive, than vice versa. Since the existence of the indicative indirect question outside of the formal prose-writings of Cicero is certain, it seems arbitrary to emend the few instances within those writings. Schmalz does not go far enough, it seems to me, when he remarks, p. 516, that editors should retain indicative indirect ques- tions "in all den Schriften, auch der klassischen Zeit, welche der Volkssprache nahe stehen." Cf. the similar remark in Kiihner- Stegmann II, §227, 6 b. Indicative indirect questions should, generally speaking, be retained wherever they occur. There is no reason, apart from tradition, for objecting to this construction. Furthermore, it seems to me that, where manuscripts vary between the indicative and the subjunctive in indirect questions, the indica- tive should be preferred. The principle of the lectio difficilior is to be followed, and, unless the manuscript tradition very markedly favors the subjunctive reading, the indicative reading is to be adopted. So I should keep the indicative in Cic, De Orat. II 60, 243,* where Wilkins, following 32, prints quotes sint: corrected back to the indicative, so that we find in the manuscript intellegaS. There is a similar instance in Cic. Tusc. I 13, 29, in cod. G (cf. p. 26). The fact that the latter example appears in our manuscripts of Saint Augustine with the subjunctive (cf. p. 26) and that an indicative indirect question in Virgil, Georg. I 57 is quoted with the subjunctive in our Seneca manuscripts (cf. p. 40) seems to be a further illus- tration of the tendency to substitute the more usual for the less usual construction. This is true whether the misquotations are the fault of scribes or of Augustine and Seneca themselves. ' For the examples cf. p. 160, note 6. * The clause may conceivably, but less naturally, be interpreted as relative. Disposal of the Indirect Question by Emendation 109 "Ergo haec duo genera sunt eius ridiculi, quod in re positumst: quae sunt propria perpetuarum facetiarum, in quibus describuntur hominum mores et ita effinguntur, ut aut re narrata aliqua quotes sunt intellegantur aut imitatione breviter iniecta in aliquo insigni ad irridendum vitio reperiantur." In Manil. IV 919 I should, similarly, read the indicative: "Atque adeo faciem caeli non invidet orbi Ipse deus vultusque suos corpusque recludit Semper volvendo seque ipsum inculcat et offert, Ut bene cognosci possit doceatque videntis, Qualis erat cogatque suas attendere leges." erat mss. except m; eat m, edd. Erat is an instance of the -use of the imperfect tense to express the discovery of a state of affairs existing before (cf. Hale-Buck, §486, 1). To be sure, there is a large number of passages in which some manuscripts have the indicative and others the subjunctive in indirect questions, but in which the subjunctive is to be preferred. See Appendix II. PART II-A SUPPLEMENTARY STUDY OF THE CONSTRUCTION IN LATER LATIN CHAPTER I Indeterminate Examples: Indirect or Direct Questions I. Questions whose Interpretation is Indifferent I. Ask 1. Quid. Petron. 55, 3: Trimalchio 'rogo' inquit 'magister, quid putas inter Ciceronem et Publilium interesse.'* II. Inform /. Quid. Martial III 30, 2: Sportula nulla datur; gratis con viva recumbis: Die mihi quid Romae, Gargiliane fads. Unde tibi togula est et fuscae pensio cellae? Unde datur quadrans? Unde vir es Chiones?^ Tert. Car. Chr. 3 (p. 894 Oehler) : Quid tanti fuit edoce, quod sciens Christus quid esset id se quod non erat exhiberet? 6. Qua Causa, etc. Petron. 67, 2: Sed narra tu mihi, Gaie, rogo, Fortunata quare non recumbit. 18. Quando. Martial V, 58, 2 : Cras te victurum, eras dieis, Postume, semper. Die mihi cras istud, Postume, quando venit. Quam longe cras istud, ubi est? Aut unde petendum? Numquid apud Parthos Armeniosque latet? • putes L O. ' die] quid C *. 110 Questions Naturally Interpreted as Indirect 111 II. Questions Which are More Naturally Interpreted AS Indirect than as Direct 1. Questions More Naturally Interpreted as Indirect, than as Direct and Not Quoted. I. Ask 1. Quid. Rutil. Lup. II 6 : Haec si praesens agit, ut dixi, res publica, quid animi estis habituri quaere' 4. Quae, etc. Enno. 262, 23: Quaero cui me servavit usus ille frugalior, si aequi observantia patrum meretur offensas. II. Inform 1. Quid. Greg. Virt. S. lul. 33(p. 578, 8): Quid de eius reliquiis in Oriente fidelium fratrum relatio signat, edicam. Greg. Virt. S. Mart. II 32 (p. 620, 29) : Quid ergo nuper actum est, multos in testi- monium exhibens declarabo.* Enno. 206, 36: Die . . . quid praeter te spei erat residuum. 2. Quis. Sen. Agam. 414: Effare casus quis rates hausit meas, Aut quae maris fortuna dispulerit duces.^ Leo, in his note on this passage. Vol. I, p. 93 of his edition, remarks that quis hausit is clearly an indirect question. Cer- tainly it would be extremely forced to consider this clause as a direct question followed by a shift to an indirect question (cf. p. 11, criterion 5). 4. Quae, etc. Asconius, In Cor- nelianam 56: Res autem tota se sic habet: in qua quidem illud primum explicandum est, de quo Metello hoc dicit.^ ' R. Stephanus changes to subjunctive. ' quod 2, p. I" quis fare nostras (vestras i^) hauserit casus rates A. disptderat Bothe. • dicai Kiessling-Schoell. 112 The Indicative Indirect Question in Latin It is not natural to interpret: "This must first be explained: concerning what Metellus did he say this?" Enno. 246, 1: Sine dissimulatione docete quae gejitis.' 13. Cur. Martial III 95, 3: Numquam dicis have, sed reddis, Naevole, semper, Quod prior et corvus dicere saepe solet. Cur hoc expectas a me, rogo, Naevole, dicas: Nam puto nee melior, Naevole, nee prior es.' 25. Num. Gellius 18, 7, 2: Quaeso, inquit, te, magister, dicas mihi num erravi quod cum vellem 5r;/i7j7opias Latine dicere, contiones dixi. 30. Various Connectives. Greg. Virt. S. Mart. II 40 (p. 624, 3): Cui ille: "Die mihi, quaeso, domine, quis es vel quod est nomen tuum." Inscriptiones His- paniae Chris- tianae (Huebner) 219 (Martin, p. 34) : Quis quantusve fuit Samson . . . Personat Esperio. Personat seems to demand the interrogative clause as its object. It would be practically impossible to interpret the question as direct. III. Find Out 30. Various Connectives. Persius III 67: Disci te et, O miseri, causas cognoscite rerum: Quid sumus et quidnam victuri gignimur, ordo Quis datus aut metae qua mollis flexus et unde, Quis modus argento, quid fas optare, quid asper Utile nummus habet, patriae carisque propinquis Quantum elargiri deceat, quern te deus esse lussit et humana qua parte locatus es in re. ' gerelis B. ' exspectas A» C«: exspectes B». Questions Naturally Interpreted as Indirect 113 It would be extremely forced to regard the indicative clauses as anything else than indirect questions. IIIB. Observe 27. An. Jer. In Gal. Ill ad 5, 13: Diligenter attende an ex consequentibus sensus iste texitur.' HID. Determine, Judge 26-27. Utrum . . . an. Aug. Civ. Dei. VIII 3: Non mihi autem videtur posse ad liquidum coUigi, utrum Socrates, ut hoc faceret, taedio rerum obscurarum ... ad aliquid apertum . . . reperiendum animum intendit ... an vero . . . nolehat immundos terrenis cupiditatibus animos se extendere in divina conari.^" HIE. Consider, Replect 23. Ut. Lucan II 682: Pompeius tellure nova conpressa profundi Ora videns curis animum mordacibus angit, Ut reserat pelagus spargatque per aequora bellum.^^ The M<-clause is ncrt; a question of fact, but„a question of deliberation. If the reading of Mi, is correct, the indicative and the subjunctive are used side by side in the same construc- tion. Probably the clause is an indirect question of deliberation, depending, upon angit. It would be less natural to interpret it as a direct question: "How shall he . . .?" For indicative questions of deliberation, cf. the comment on Plant. Epid. 274 (p. 12). " var. texatur. ^^ intenderit p v, Dombart: inienderet C q: intendit A K F and the othei codd. " reserat Mi: referat Zi: reseret other codd., Hosius. 114 The Indicative Indirect Question in Latin IV. Know 1. Quid. Jer. C. Vig. 5: Quid est "illud nescio quid" scire desidero. Aug. Conf. 1 13, 20: Quid autem erat causae cur Graecas litteras oderam quibus puerulus imbuebar, ne nunc quidem mihi satis exploratum est. 2. Quis. Lucan I 126: Nee quemquam iam ferre potest Caesarve priorem Pompeiusve parem. Quis iustius induit arma, Scire nefas; magno se iudice quisque tuetur: Victrix causa deis placuit, sed victa Catoni. 26-27. Utrum . . . an. Amm. Mar.XVIll, 12: Quae utrum ut vanus gerebat, ... an mandatu principis . . . nefanda multa tempta- bat . . . latuit. 27. An. Sen. Nat. Quaest. V 18, 12: Unde scio an nunc aliquis magnae gentis in abdito dominus . . . non contineat intra terminos arma, an paret classes ignota moliens? Unde scio, hie mihi an ille ventus bellum in- vehet?'-^ Apul. Met. VI 5, 391:, Qui scias an etiam quem diu quaeritas illic in domo matris repperies?^^ Greg. Hist. Fr. II 9 (p. 74, 15): Cum autem eos regales vocet, nescimus, utrum reges fuerint, an in vices tenuerunt regnum." In the last seven passages it is highly improbable that the questions are direct. ^ invehat Madvig, Gercke. " repperies F: reperias v, repperias Helm. " tenuerinf A 1 C 1. Questions Naturally Interpreted as Indirect 115 V. Hear I. Quid. Sen. Dial. VII 25, 1: Quid ergo est quare illas non in bonis nume- rem et quid praestem in illis aliud quam vos, quoniam inter utrosque convenit habendas, audite." Haase places an interrogation-point after est. Greg. Hist. Fr.V 43 (p. 234, 26) : Audi quid lesus deus noster, cum ad resus- citandum venit Lazarum ait.'* Hist. Fr., VI 40 (p. 280, 15): Absculta quid e caelis loquitur. VI. See II. Quot. Claud. Mam. De Statu Animae, Epilogus (p. 192, 22): Nunc etiam videamus quot genera sunt corporum. 26. Utrum. Vegetius IV 3, 9: Quod utrum bene opinantur usus viderit.^' It would be very forced to interpret this question as direct. 27. An. Tert. Virg. Vel. 9: Videamus nunc an . . . praescripta de muliere in virginem spectant.^' 2. Questions More Naturally Interpreted as Indirect, than as Directly Quoted. I. Ask 1. Quid. Sen. Dial I 1, 1: Quaesisti a me, Lucili, quid ita, si pro- videntia mundus regeretur, multa bonis viris mala accidere.^' " est codd. : sit Lipsius, Vahlen. " quod, changed by the same hand to quid, B 2. " opinentur Gesner, Schneider, Lommatzsch. " spectent codd. tres Vatican!. " acciderent E.i 116 The Indicative Indirect Question in Latin Lucan IX 563 : Tua pectora sacra Voce reple; durae saltern virtutis amator Quaere quid est virtus, et posce exemplar honesti. Enno. 232, 20: Caelestis dispensatio facienda suggerit, si quid amor optat inquiras. Buecheler 1136, 1 (C I L VI 9693) : Quid sibi volt, quaeris, tellus congesta, viator? 2. Quis Greg. Hist. Fr. VIII 13 (p. 333, 4):^° Ait: 'Patruus tuus, o rex, diligenter inter- rogat, quis te ab hac promissione retraxit, ut sacerdotes regni vestri ad concilium, quod simul decreveratis, venire differrent.'^^ 4. Quae, etc. Sen. Ep. Mor. VI 6, 18: Tertium genus est eorum quae proprie sunt: innumerabilia haec sunt, sed extra nostrum posita conspectum. Quae sunt interrogas? Propria Platonis supellex est.^^ Buecheler 222, 7 (C IL X 4183; a very late inscrip- tion) : Si nomen queres, qui leges, Mensem priorem cogita. Si qui fecerunt queritas, Parentes dixi, sufficit. Greg. Hist. Fr. X 16 (p. 427, 13) :2'' Exemplar ludicii . . . evocatis partibus, interrogata Chrodieldis vel Basina, quare tam audacter contra suam regulam, foribus monas- terii confractis, discesseranf, et hac occasione congregatio adunata discessa sit.^' '" For the shift in person cf. p. 25, criterion 4. ^ traxit A 1. 22 sint P Pr. a r B. 23 discesserint A 1, D 5; discessirent B 1. Questions Naturally Interpreted as Indirect 117 10. Quantus, etc. Martial VI 88:"^ Mane salutavi vero te nomine casu Nee dixi dominum, Caeciliane, meum. Quanti libertas constat mihi tanta requiris? Centum quadrantes abstulit ilia mihi.^^ 12. Quotus, etc. Lucan VIII 170: Rectoremque ratis de cunctis consulit astris, Unde notet terras, quae sit mensura secandi Aequoris in caelo, Syriam quo sidere servet, Aut quotus in plaustro Libyam bene dirigit ignis.''^ It would be extremely forced to understand this example as showing a shift from indirect to direct quotation. 23. Ut. Val. Flac. VII 120: Quaerit ut Aeaeis hospes consederit oris Phrixis, ut aligeri Circen rapuere dracones. The parallel use of a subjunctive indirect question favors the interpretation of the indicative clause as an indirect question. Furthermore, a directly quoted question, "How did the dragons carry Circe away"? would be absurdly naive. 27. An. Martial III 32, 1:^' An possum vetulam quaeris, Matronia: possum Et vetulam, sed tu mortua, non vetula es.^* Aug. Serm. LXVI, 3:" Quaeris an ego sum? ,, „ VIII. Wonder 13. Cur. Stat. Theb. IV 333: Sunt omina veri: Mirabar, cur templa mihi tremuisse Dianae Nuper et inferior voltu dea visa, sacrisque Exuviae cecidere tholis; hoc segnior arcus Difficilesque manus et nuUo in vulnere certae. It would seem even more harsh to interpret the last example as a relative clause than as a directly quoted question. " For the shift in person cf. p. 25, criterion 4. ''Lindsay n. cr.: "constat A»: constet B& (sed constat L ante corr.): constel (E A C G) vel constat (X B) C»." ^ dirigit Mi, V Zi; dirigai other codd., Hosius. " For the shift in person cf. p. 25, criterion 4. >8 Thus A»: non possum B» C». Edd. read subj. or use quotation marks. CHAPTER II Indeterminate Examples : Indirect Questions or Exclamations I. Clauses whose Interpretation is Indifferent VI. See 30. Various Connectives. Stat. Theb. I 188: Cernis ut erectum torva sub fronte minetur Saevior adsurgens dempto consorte potestas, Quas gerit ore minas, quanto ptemit omnia fastu. Hicne umquam privatus erit? It is possible that a new sentence, an exclamation, begins with quas. II. Clauses Which are More Naturally Interpreted AS Indirect Questions than as Exclamations II. Inform 4. Quae, etc. Stat. Theb. XII 805: Non ego, centena si quis mea pectora laxet Voce deus, tot busta simul vulgique ducumque. Tot pariter gemitus dignis conatibus aequem: Turbine quo sese caris impleverit audax Ignibus Euhadne fulmenque in pectore magno Quaesierit; quo more iacens super oscula saevi Corporis infelix excuset Tydea coniunx; Ut saevos narret vigiles Argia sorori; Arcada quo planctu genetrix Erymanthea cla- Arcada, consumpto servantem sanguine vultus, Arcada, quem geminae pariter Severe cohortes.^ It would be possible to begin a new sentence, an exclamation, at line 805, but it seems more natural to regard this verse as an indirect question. » clamat P: clam'^ B1Q2; clamet other codd., Klotz. 118 Clauses Naturally Interpreted as Indirect Questions 119 10. Quantus, etc. Greg. Vit. Patr. VI (7) (p. 685, 21): lam vero in exsequiis eius quantus planctus, quanti populi adfuere, enarrari vix potest. Other examples are: Vit. Patr. VIII (10) (p. 700, 5): Virt. S. Mart. I 36 (p. 605, 25). Enno. 107, 4: Quantum tunc, admirande pontifex, tua plus egit absentia, quantum imperavit humilitas deprehensa, dicant illi quos de exulibus ditis- sinios reddidisti. III. Find Out 10. Quantus, etc. Enno, 106, 26: Quantum acutior fuit verborum quam ferri lammina, hinc, lector, agnosce. Enno. 129, 21: Quantum apud me pondus est perlatoris advertite. 30. Various Connectives. Greg. Hist. Fr. VI 40 (p. 280, 1) : Nam tu, qui Paulo apostolo derogas, et • sensum eius non intellegis, percipe quam caute loquitur, et iuxta ut recepere quis potest, averte qualiter praedicat inter gentes incredulas, ut nuUo onos grave videatur imponere. HID. Determine, Judge 1. Quid. Greg. Virt. S. Mart. I 22 (p. 600, 8) : Deliberataque ab omni inpedimento lingua, beati Martini miraculum populis testabatur, dicens: "Ecce, qmd in hac nocte sanctus Dei operatus est, me teste, probate!' HIE. Consider, Reflect 8. Qualis, etc. Greg. Hist. Fr. X 1 (p. 407, 26) : Pensate ergo, qualis a conspectu district! ludicis pervenit, cui non vacat fiere quod fecit.* ' teniat A 1. 120 The Indicative Indirect Question in Latin 9. Qualiter. Greg. Glor. Conf. 20 (p. 760, 4) : Et praesertim rememoramini librum vitae beati Martini et recolite qualiter verba sacrata promenti de capite globus ignis egressus, usque ad caelos visus est conscendisse. It would be very forced to interpret the last two examples as exclamations. 30. Various Connectives. Marcus Aurelius (Fronto, Ad M. Caesarem et in- vicem III 19, p. 56, 16): Qualem mihi animum esse existimas, quom cogito quam diu te non vidi, et quam ob rem non vidi? A. Ebert, "De Syntaxi Frontoniana," Acta Seminarii Philo- logici Erlangensis II (1881), p. 347, says: "Interrogatio directa, ubi ratio interrogationem indirectam cum coniuaictivo desiderat." He cites two passages as parallel. The one, Ad Varum I 1, p. 114, 22, by Fronto, is probably a direct question; the other, by Verus, is cited on p. 157. It seems very forced either to interpret the clause under consideration as Ebert does, or to regard it as a quoted exclamation. The natural interpretation is that the clause is an indirect question. V. Hear 1. Quid. Claud. Mam. De Statu Animae 1, 6 (p. 42, 12): Saepe dicimus: vide quid sonat, cum sonus ad auditum pertineat, non ad visum, nee tamen ita dicimus audi quid lucet. 1. Quid. VI. See Claud. Mam. De Statu Animae 1, 6 (p. 42, 11), just cited. Clauses Naturally Interpreted as Indirect Questions 121 Greg. Hist. Fr. VI 40 (p. 280, 7): Adverte potius cautelam eius et vide astu- tiam, quid aliis dicit quos robustiores videbat in fide.' Greg. Glor. Mart. 105 (p. 561, 6): Tamen vide quid agit idem doctor.^ Greg. Virt. S. Mart. I 28 (p. 602, 3): Ecce quid, sancte, praestas fidelibus, qui tua moenia expetunt proprie. 4. Quae, etc. Stat. Theb. XII 260: Cernis quo praedita cultu. Qua stipata manu, iuxta tua limina primum Oedipodis magni venio nurus? Buecheler 457, 2 (CIL XI 4311): Tu quicumque legi[s ti]tulum nostrum nomen- que requiris Aspice quo fato raptus mihi spiritus or[e] est. Enno. 234, 21: Ecce vide qua mecum arte contendis. 9. Qualiter. Com. Instr. I 31, 1 Intuite dicta Salomonis .... . . . qualiter vos ille detractat. Greg. Glor. Conf. 49 (p. 777, 20): Ait: "Vae mihi, qui in amaritudine cordis maledixi huic arbori. Ecce enim qualiter aruit." 10. Quantus, etc. Juv. V 67: Ecce alius quanto porrexit murmure panem Vix fractum, solidae iam mucida frusta farinae. Quae genuinum agitent, non admittentia mor- sum. Greg. Glor. Mart. 105 (p. 560, 31): Vides ergo quantum distal inter conversa- tionem caelestem et opolentiam saecularem, quantum distal inter martyrum divitias et saeculi pompas; vides qualia martyribus sint conlata praemia ad vitae religiosae conpendia. ' dicit om. A 1. 'Thus 3. 122 The Indicative Indirect Question in Latin Greg. Mirac. B. Andr. Apost. 23 (p. 839, 29) : "Videte, dilectissimi, quantum praevalet ini- micus, nam Trofime propter pudicitiam dam- naverunt scorto."* Cor. lust. I 181: Aspice quanta fuit nostrae simul urbis et orbis Provida cura seni. 11. Quot. Sil. It. Pun. II 348: Indole non adeo segni sumus. Aspice, turmae Quot Libycae certant annos anteire labore Et nudis bellantur equis. Ipse, aspice, ductor. Cum primam tenero vocem proferret ab ore, lam bella et lituos ac flammis urere gentem lurabat Phrygiam atque animo patria arma movebat. 19. Quam. Sen. De Ben. II 29, 1: Vide quam iniqui sunt divinorum munerum aestimatores et quidem professi sapientiam.' Sen. Here. Fur. 1299: Ecce quam miserum metu Cor palpitat pectusque sollicitum ferit. Buecheler 1489 (CIL II 4426): Aspice quam subito marcet quod floruit ante, Aspice quam subito quod stetit ante cadit. Nascentes morimur finisque ab origine pendet. Buecheler 1839, Rossi museo Pio- Lateranense ta- bula 23 : Respice quam parbus cubat hie sine f elle palum- bus Dum luce est ista frunitus. ' praevalet 1 (3). ' Thus codd., Hosius: sint earliei edd. Clauses Naturally Interpreted as Indirect Questions 123 Claud. Mam. De Statu Animae I 9 (p. 49, 18) : Cernis quam facile indoctae figmentum praesumptionis evanuit, praeterquam innumera super hoc argumentari promptissimum est. It would be extremely harsh to interpret thjs quam-dause otherwise than as an indirect question. Enno. 197, 3: Sed vide per rerum providentiam quam cauta est seniorum dispensatio et fabricatis plena sermonibus. 23. Ut. Sil. Ital. XIII 446: Interea cerne ut gressus inhumata citatos Fert umbra etproperat tecum coniungere dicta.'" Buecheler 369, 1 Cernis ut orba meis, hospes, monumenta locavi Et tristis senior natos miseranda requiro. 30. Various Connectives. Apul. Met. V9, 337: Vidisti, soror, quanta in domo iacent et qualia monilia, quae praenitent vestes, quae splendicant gemmae, quantum praeterea passim calcatur aurum. It would be extremely harsh to interpret quanta . . . aurum otherwise than as indirect questions. TertuUian, Ad Uxo- rem I 8: Vide quam ex aequo habetur qui viduae benefecerit, quanti est vidua ipsa, cuius assertor cum domino disputabit.' Greg. Glor. Conf. 96 (p. 809, 20) : Ecce quales quantasque suis congregat divitias mundi paupertas, ut eis non solum quae volue- rint Redemptor, qui cuncta ex nihilo condidit, tribuat, verum etiam ipsa eis elementa iubeat famulari. ««/ert OV: «< LF. ' habeatur BC. 124 The Indicative Indirect Question in Latin Glor. Conf. 108 (p. 818, 17): Ecce quid tribuit elymosina. Ecce quales thesauros Sanctis suis, qui se in pauperibus dilegunt, Deus indulget. VIII. Wonder 9. Qualiter. Greg. Vit. Patr. VII (4) (p. 689, 27): Admirabile est enim et illud miraculum, qualiter beatum corpus eius, cum post multa tempora transferretur, apparuit gloriosum. Greg. De Cursu Stellarum 16 (p. 863, 5): Septimum est enim miraculum, qualiter luna in ter quinis diebus vel crescit ad integrita- tem vel ad exiguitate minuitur.* It would be very forced to interpret the last two examples otherwise than as indirect questions. 'minuitur 2 (3): minuatur 1. CHAPTER III Indeterminate Examples: Indirect Questions or Relative Clauses I. Clauses whose Interpretation is Indifferent II. Inform 4. Quae, etc. Apul. Met. IS, 24: Quod ibidem passim per era populi sermo iactetur quae palam gesta sunt. Helm changes to iactet. Prescott in Class. Phil. X 358 says that "iactetur is quite possible with ellipsis of de eis." On Prescott's interpretation the quae-clause is relative. It might equally well, it seems to me, be considered an indirect question. Cf. the subjunctive indirect question in Lucan IV 201, cited p. 128. Apul. Met. Ill 3, 178: Rem denique ipsam et quae nocte gesta sunt cum fide proferam. Met. VIII 1, 505 : Sed ut cuncta noritis, referam vobis a capite, quae gesta sunt quaeque possent merito doc- tiores, quibus stilos fortuna subministrat, in historiae specimen chartis involvere. Met. VIII 14, 545: Et enarratis ordine singulis, quae sibi per somnium nuntiaverat maritus quoque astu Thrasyllum inductum petisset, ferro sub papil- 1am dexteram transadacto corruit. Helm, who punctuates thus, apparently considers the quae- clause an indirect question. It would be possible to punctuate: "Singulis quae . . . maritus, quoque . . .," and to interpret: "When everything that her husband reported had been related, and it had been related by what craft she had enticed Thrasyllus and attacked him. . . ." Apul. Met. XI 25, 808: Nee mihi vocis ubertas ad dicenda quae de tua maiestate sentio, sufficit. 125 126 The Indicative Indirect Question in Latin Apul. Apol. 42 : Memini me apud Varronem . . . hoc etiam legere: Trallibus de eventu Mithridatici belli magica percontatione consultantibus puerum in aqua simulacrum Mercuri contemplantem quae futura erant CLX versibus cecinisse. Asconius, In Milo- nianam 39: Cognovi . . . Munatium in contione ex- posuisse populo quae pridie acta erant in senatu. 6. Qua Causa, etc. Ter. Andr., Alter Exitus Supposi- ticius 6: Id quam ob rem non volui eloquar. HIE. Consider, Reflect 4. Quae, etc. Apul. Met. Ill 13, 196: Abiectus in lectulo meo, quae gesta fuerant, singula maestus recordabar. IV. Know 4. Quae, etc. Petron, 8, 2: Si scires, inquit, quae mihi acciderunt. V. Hear 4. Quae, etc. Statius Theb. V 734: Audite, o ductor Nemeae lectique potentes Inachidae, quae certus agi manifestat Apollo. Amm. Mar. XIV 11, IS: Nullam videndi vel audiendi quae ferebant furari potuit facultatem. Greg. Mirac. B. Andr. Apost. 24 (p. 840, 40) : Nunc autem, si audierit quae mihi con- tigerunt, blasphemat Deum tuum. Clauses Naturally Interpreted as Indirect Questions 127 VI. See 4. Quae, etc. Sen. De Clem. 1 13, 3; Qui ubi circumspexit quaeque fecit quaeque facturus est et conscientiam suam plenam sceleribus ac tormentis adaperuit, saepe mortem timet, saepius optat, invisior sibi quam ser- vientibus.^ II. Clauses Which are More Naturally Interpreted AS Indirect Questions than as Relative Clauses I. Ask 6. Qua Causa, etc. Sen. Controversiae II 5, 14: Quaeris quare non peperit? 7. Quo Modo, etc. Jer. In Naum . ad 2, 1 : Quaeritur quomodo eversa est iniuria lacob. 13. Cur. Enno. 21, 20: Si quaeras cur silentio vestro multata non reticet prodiga frons pudoris. 21. Unde. Hermerus Mulom. Chironis 191: Item rationem huius morbi, unde nascitur, inquirere debemus. II. Inform 1. Quid.^ Petron. 76, 11: Tantum quod mihi non dixerat quid pridie cenaveram. Priscian, Keil II, p. 421, 20: Indicativus, quo indicamus vel definimus, quid agitur a nobis, vel ab aliis. Priscian, Keil II, p. 423, 25: Indicativus autem dicitur quod per eum indicamus quid agitur. » quae fecit S: om. A. 'For the examples with quid cf . App. 200. 128 The Indicative Indirect Question in Latin Greg. Glor. Conf. 8 (p. 753, 25) : Velim, o sonipes, si tibi os Dominus, uti quondam fecit asinae, reseraret, diceris, quid vidisti spectabile, ut non incederes. 4. Quae, etc. Quint. Inst. IV 5, 26: Et divisa autem et simplex propositio . . debet esse . . . brevis nee uUo supervacuo onerata verbo; non enim quid dicamus, sed de quo dicturi sumus ostendimus.' Lucan IV 201: Extrahit insomnis bellorum fabula noctes, Quo primum steterant campo, quae lancea dextra Exierit.* For the possibility that the g'Mo-clause is relative, cf. p. 125. The parallel use of the subjunctive clause, quae . . . exierit renders it almost certain that the indicative clause is an indirect question. Gellius 9, 2, 5: Quaeso autem te, cum bona venia dicas mihi quibus nos uti posse argumentis existimas, ut esse te philosophum noscitemus. Greg. Hist. Fr. IX 2 (p. 359, 22): Quae autem ibi ipsa die virtutes apparue- runt, vel qualiter fuerit funerata, in libro miraculorum plenius scribere studui. Greg. Mirac. B. Andr. Apost. 22 (p. 838, 35): "Surge et enarra nobis quae tibi contigit."^ Enno. 109, 4: Quae ibi fuerunt flumina lacrimarum, quanti planctus, silebo. Enno. 270, 17: Ostendi post oris feirias quae tibi per aurium callem divitiae commearunt. Cor. lo. VIII 207: Docet quae prima salutis Causa fuit. ' sumus A B : sinms other codd., Radermacher. * steterant Mi VPUi; stetennt other codd., edd. • contigerint 1, 2, S: conligerunt 3. Clauses Naturally Interpreted as Indirect Questions 129 Aetheria Peregrina- tio XII (p. 18, 11): "Si vultis videre loca, quae scripta sunt in libris Moysi, accedite foras hostium eccle^ae et de summitate ipsa, ex parte tamen ut possunt hinc parere, attendite et videte. Et dicimus uobis singula, quae sunt loca haec, quae parent." 5. Qui (Masculine Singular Substantive; cf. p. 56, n. 29). PUny N.H. XXXVII (38) 119: Adscribitur et qui primus tinxit. 6. Qua Causa, etc. Amm. Mar. XIV 10, 13: Ut . . . ostendam qua ex causa omnes vos simul adesse volui . . . accipite . . . quae succinctius explicabo. Amm. Mar. XIV 11, 21: Compulsuri eum singillatim docere quam ob causam quemque . . . necatorum iusserat trucidari. Buecheler 932, 2 (Notizie d. Scavi 1882,.p.436 . . . Pompeis in pos- ticae pariete.) : Miximus in lecto. Fateor, peccavimus, hospes. Si dices quare nulla matella fuit. Hermerus Mulom. Chironis 352: Cuius et rationem ego protinus reddam, quare . . . insanabilis est. 7. Quo Mode, etc. Sen. Suasoriae 1 10: Hie dixit incerta peti . . .; hie matrem de qua dixit quomodo ilia trepidavit etiam quom Granicum transiturus esset.* Sen. Ep. Mor. XX 4, 21 : Si tamen exigis, dicam quomodo omne animal perniciosa intellegere conatur. ' Adolph Kiessling (ed. Leipzig, 1872) puts: after de qua dixit, and adopts W. Mueller's change of esset to esses, thus interpreting the clause as a directly quoted exclamation. 130 The Indicative Indirect Question in Latin Columella IX 3, 4: Nam saepe morbis intercipiuntur, quibus quemadmodum mederi oportet suo loco dicetur. Tert. Adv. Praxean 2 end: Quo modo numerum sine divisione patiuntur procedentes tractatus demonstrabant. Vopiscus, Hist. Aug. XXIX 8, 9: Nihil illis opto, nisi ut suis puUis alantur, quos quern ad modum fecundant, pudet dicere. Aetheria V (p. 10, 14) : Ostenderunt etiam quemadmodum per ipsam vallem unusquisque eorum abitationes habuerant. Aetheria VIII (p. 14, 16): Dixit nomen ipsius arboris quemadmodum appellant earn Grece. Aetheria XLV (p. 52, 6) : Et illud etiam scribere debui, quemadmodum docentur hi qui baptidiantur per pascha. For the last example especially, owing to the illud, the inter- pretation of the dependent clause as interrogative is almost inevitable. Hermerus Mulom. Chironis 38: Qua ratione singula vitia inuri debent, primum demonstro. 10. Quantus etc. Apul. Met. II 22, 145: Nee satis quisquam definire poterit quantas latebras nequissimae mulieres pro lubidine sua comminiscuntur. Enno. 113, 5: Qui non potest quantus est una tractatuum forma monstrari. For these two examples, compare p. 55, n. 27. In the Apul. passage, the satis makes it especially difficult to supply, in thought, tantas. 11. Quot. Asconius, In Cor- neUanam, 62 (p. ^ , , , >,■, ^\. Quo loco enumerat, cum lex feratur, quot ' ' loca intercessionis sunt.' ^ sunt codd.: sint Sigonius, Clark. Clauses Naturally Interpreted as Indirect Questions 131 20. Ubi. Enno. 209, 32 : Taceo ubi tibi iniuncta est pax diuturna. 21. Unde. Aug. Conf. V 8, 14: Et hoc unde mihi persuasum est, non prae- teribo confiteri tibi. Amm. Mar.XIX4, 1: Quae genera morborum unde oriri solent breviter explicabo. 30. Various Connectives. Lucilius Aetna 220: Nunc quoniam in promptu est operis natura solique, Unde ipsi venti, quae res incendia pascit, Cur subito cohibentur, iners quae causa silenti, Subsequar.* Vopiscus, Hist. Aug. XXVI, 26, 4 (a letter of Aurelian) : Dici non potest quantum Mc sagittarum est, qui belli apparatus, quantum telorum, quantum lapidum. For the possibility that the quantum-clsLUse is relative, see p. 55, n. 27. IIB. Discuss 1. Quid.^ S. C. Vellaeanum a. 46 (e Dig.), Bruns p. 194: Quod M. Silanus et Vellaeus tutor cos. verba fecerunt de obligationibus feminarum quae pro aliis reae fierent, quid de ea re fieri oportet, de ea re ita censuere. III. Find Out 1. Quid. Amm. Mar. XIV 10, 15 : Tamquam arbitros vos quid suadetis operior. Amm. Mar. XVI 12, 18: Experieris quid miles . . . efficiet. ' pascat Matthiae. cur G: cum the other codd. » For the possibility that the gwid-clauses aie relative, cf. App. p. 200. 132 The Indicative Indirect Question in Latin Greg. Mirac. B. Andr. Apost. 24 (p. 840, 18): Hermerus, Mulom. Chironis 984: Enno. 275, 24: Enno. 291, 22: 4. Quae, etc. Quint. Inst. IV 5,1: Huebner, Inscrip- tiones Hispaniae Christianae 255 (Martin p. 34) : Greg. Hist. Fr. Ill 31 (p. 135, 12): Cor. lo. I 103: Oportet hunc resuscitari, ut cognoscamus quid in eum adversarius est operatus.^" Si quod iumentum . . . doluerit, diligen- tius eum tractato, ut scias quid ei dolet. Rogo ut quid circa te agitur tabellarum promulgatione cognoscam.^^ Rogo ut quid in causa vestra de Roma nuntiatur agnoscam.'^ Quod ea fiat causa lucidior et index attentior ac docilior, si scierit et de quo dicimus et de quo postea dicturi simus.'^' Omnis conventus . . . veneratur reliquias. conoscat q(u)orum Hie verc cum dedicisset quae meretrix ista commiserat, qualiter propter servum, quern acceperat, in matrem extiterit parricida. . . ." Noscere quae saevi fuerant discrimina belli. IIIB. Observe 1. Quid}^ Festus 219 (Lindsay p. 208): Observasse dicitur qui observat quid cuius- que causa facere debet.^^ 10 qiiae 2, 3, 4a. " agatur Sirmond (ed. 1611). ^ nuntietur Sirmond. "Thus A(Baehreiis, "Beitrage," p. 525, says correctly; he calls it "Variatio beim Rel.") : sumus the other mss., Radermacher. ^* cummiserat B2. exUterat Al, CI: extitirii Bl: extiteret B2: sHterel B4. " For the examples with quid cf. App. p. 200: " debet W: debeat X, ed. princ, Lindsay. Clauses Naturally Interpreted as Indirect Questions 133 HID. Determine, Judge 7. Quo Modo, etc. Jer. Ep. 22, 10: Ad exemplum horum poteris tibi ipsa col- ligere quomodo primus de paradise homo . . . in banc lacrimarum deiectus est vallem. 18. Quando. Terentius Scaurus in Varro, Reliquo- rum de Gramma- tica Librorum Fragm., p. 223 (Keil Gram. Lat. VII 29, 3): Quando ergo ex, quando e dici oportet, ex consequentibus vocabulis animadvertitur.^' 30. Various Connectives. Tert. Praescr. 22 (end): Sed nee ecclesiam se dicant defendere, qui quando et quibus incunabulis institutum est hoc corpus probare non habent. Aug. Civ. Dei V 12 (Dombart 214, 22): Unde intellegi potest quern finem volebant esse virtutis et quo earn referebant qui boni erant, ad honorem scilicet. HIE. Consider, Reelect 13. Cur. Apul. Apol. 16: Etiam ilia ratiocinatio necessaria est, cur in planis quidem speculis fexme pares optutus et imagines videantur, in tumidis vero et globosis omnia defectiora, at contra in cavis auctiora; ubi et cur laeva cum dexteris permutentur; quando se imago eodem speculo turn recondat penitus, turn foras exerat; cur cava specula, si exadversum soli retineantur, appositum fomi- tem accendunt; qui fiat ut. . . ^^ " oportet B (the most trustworthy cod.; cf. Keil VII, p. 6) : oporteat P, edd. '' vUf^nt (em. ead. m.) F. accendunt F(p: accendant Bosscha, Helm. 134 The Indicative Indirect Question in Latin Cur . . . accendunt is naturally interpreted as an indirect question, like the subjunctive clauses preceding and following it. 15. Quo. Sen. Dial. I 4, 4: Avida est periculi virtus, et quo tendit, non quid passura sit, cogitat, quoniam etiam quod passura est gloriae par est.*' 18. Quando. Tert. Praescr. 8 (beginning): Quando banc vocem, dominus emisit, re- cordemur. 20. Ubi. Com. Instr. 1, 33, 8: Nee respicis ubi moraris. IV. Know /. Quid.^o Amm. Mar. XIV 7, 14: Quid rerum ordo postulat, ignorare dis- simulantem. . . . Aug. Serm. CCXXXVI 3 : Nescio quid ibi erit. 4. Quae, etc. Hyginus 167 (20, 16) : Ut scias quae voluptas est cum deo concumbere. This example is taken from M. Tschiassny, Studia Hyginiana (Vienna, 1888) p. 23, n. 50. Tschiassny remarks, "In interrog. obi., quarum modum recte in universum Hyginus adhibuit coniunctivum, praeter titulos fabularum quos nescio an ab eo profecti non sint, semel a recta norma declinavit," and cites this passage. Sen. Ep. Mor. 1 4, 8: Lex autem naturae, scis quos terminos nobis statuit?2i Sen. De Paupertate 4, 10: Lex autem naturae, scis quos terminos nobis statuit? 1' tendat edd. " For the examples with quid, cf. App. p. 200. " statuat R e o, Fickert. Clauses Naturally Interpreted as Indirect Questions 135 Martial IX 92, 1: Quae mala sunt domini, quae servi commoda nescis, Condyle, qui servum te gemis esse diu.^^ Apul. Met. V 31, 379: Tunc illae non ignarae, quae gesta sunt, palpare Veneris iram saevientem sic adortae. For the possibility that the dependent clause in this example is relative, cf. p. 53, criterion 3. Tert. Ad Nationes I 1: Scio plane qua responsione soletis . . . convenire. Amm. Mar. XX 8, 17: Scio . . . quas rerum acerbitates . . .Con- cordia . . . meliorem revocavit in statum. Greg. Hist. Fr. X 3 (p. 411, 13): Dicens: "Hodie apparebit, cui Divinitas obtenere victoriam praestit."^ Enno. 83, 19: Scitis optime quae a vobis et per me et ore proprio sanctus pater vester dominus episcopus postulavit. Enno. 100, 14: Scis quae te poUicebaris acturum. Enno. 211, 6: Meministis, socii, cuius ad haec loca com- meastis imperio." Hermerus Mulom. Chironis 567 : Cum nodum fecerit, regulas solves die primo L. Deinde tunc scies quo tempore nodum faciet, et unctionibus uteris donee ad sanitiem perducas. 6. Qua Causa, etc. Sen. Ep. Mor. XIX 7, 7: Scis quare non possumus ista? Quia nos posse non credimus. Haase punctuates: "Scis quare? Non possumus. . . ." Jer. In Ephes. Pr. : Per quos [sc, divinos libros] et Deum dis- cimus et quare nati sumus non ignoramus. *» sunt A» C»: sint B». " praestitit A 1: praestai corr.: praestit C 1: praestabit DS. " commeassites B': commeasseUs B*. 136 The Indicative Indirect Question in Latin CIL II 6130 (Martin p. 34) : Quare mortuus sum nescio. Hermerus Mulom. Chironis451: Curato et hoc utique scire, ex qua causa haec vitia nascuntur, quam numquam nemo scripsit.^^ 7. Quo Modo, etc. Apul. Met. 125, 78: lam enim faxo scias quem ad modum sub meo magisterio mali debent coherceri.^' Hermerus Mulom. Chironis 352: Quemadmodum intelligere quis possit, qua ratione succurrere huic vitio nemo potest. 20. Ubi. Lucan VIII 645 : Nescis, crudelis, ubi ipsa Viscera sunt Magni; properas atque ingeris ictus, Qua votum est victo.^' Com. Instr. II 15, 13 : lam scitis ubi demersit. 21. Unde. Com. Instr. I 24, 7 : Unde processisti nescis nee unde nutriris. V. Hear 7. Quo Modo, etc. Apul. Met. IX 30, 650: Accipe igitur, quem ad modum homo curio- sus iumenti faciem sustinens cuncta, quae in perniciem pistoris mei gesta sunt, cognovi. Aug. Conf. V 14, 24: Cum enim non satagerem discere quae dicebat, sed tantum quemadmodum dicebat audire . . ., veniebant in animum meum simul cum verbis, quae diligebam, res etiam, quas neglegebam. "Odern. cr.: "sciret." ™ debeani Helm, with n. cr.: "debeo (• quod second hand of B, CD: quid first hand of B. Relative (Quis) Quid 197 Poen. 1212: Ad. Quis reuocat? Ag. Qui bene uolt uobis facere.^" The preceding quis may have caused an error in copying.'*! Ter. H. T. 638: Quam bene vero abs te prospectumst Quid uoluisti cogita.^ Cic. Att. X 10, 4: Ocellam cuperem, si possem palam, quid acuone effeceram.''' Petron. 50 : Ignoscetis mihi quid dixero.^ Rutilius Lupus II 10: Hoc schema efficitur, cum quaerimus, quid aut quemadmodum pro rei dignitate dicimus, nee reperire nos ostendimus. Carm. Epigr. (Biicheler) 520, 7 (Ephem. Epigr. V 1049). (3d cent. A.D.): Et quid [non m]ulti poterant iuvene[s], hie semper solus a[gebat. Amm.Mar.XIVll, 11: Qui eum . . . proficisci pellexit vultu ad- simulato saepius replicando, quod fiagrantibus votis eum videre frater cuperet patruelis, quid per imprudentiam gestum est remissurus, ut mitis et clemens.* Venantius Fortuna- tus Vita S. Mar- tini IV 440: Ni cito subvenias, rapit omnia pestis acerba. Per spatium ferale trahens in Tartara cunctos, Et grassata diu quid agant contagia perdunt. Past. Herm. Vis. 3, 3, 3: Quid ergo nunc vis, interrogare (infinitive for imperative; from Stolz-Schmalz, p. 529). ^'quiP: quis A. " A shows a similar error in Pers. 636, where quaefuit has been copied as quifuit, owing to the presence of quifuit two verses below. " quid codd.; quod Bothe, Tyrrell. ^ So M. Editors emend to quod a Curione, " quod Muncker, Buecheler. " quid Kiessling, Clark: quod E B G. 198 The Indicative Indirect Question in Latin Greg. H. F. VI 40 (p. 280, 18) : Certe si oppilatas habeas auras, ut ista non audias, crede apostolis, quid in monte audierunt, cum transfiguratus lesus in gloria loqueretur cum Moysi et Helia.^^ Do the examples in group II establish the use of {quis) quid as a relative pronoun equivalent to qui (quod)? In weighing the evidence, we should reject those passages in which the dependent clauses may have some interrogative coloring. Thus, the Terence example may be an instance of confusion of the relative and interrogative clause, for the verb prospicere is such as might be followed by either construction. In the Rutilius passage, similarly, the clause under discussion, following, as it does, the verb quaerimus, may be an instance of confusion, or may, indeed, actually be an indirect question. In the passage from Vita S. Mart., it may well be that the quid-clause after perdere is an extension from the quid-cla.\ise after non habere and the like (cf. p. 193, n. 6). In Greg., H. F. VI 40, the quid-clause seems to be an indirect question, after the implication in cye 178-125. Ill 13, 196-126. V 9, 337-123. V 31, 379-53; 135. VI 5, 391-114; 171ii. 51. VIII 1, 505-125. VIII 14, 545-125. IX 30, 650-136. XI 2.5, 808-125. Uipl 'Ep/njm'oj, V 268-145. Asconius In Cornelianam: 54r-89; 144; 171n.51. 56-111. 62-130. 69-153. In MUonianam: 39-126. Pseudo-Asconius-176. Augustine General-172n. 52; 178. Civ. Dei.: 128-142. 116-145. V 12- 133; 137. VIII 3-113. X 5-137. XVII 20-79; 141. Conf.: I 13, 20-114. V 8, 14-131. V 9, 16-148. V 14, 24^136; 171. Serm.: II 7-156. V 7-152. XXII 9- 137. XXIV 4^155. XXIV 4^155. XXXVII 10-155. XLV 6-141; 155; 156. XLVI 9-156. LII 13-141. LVII 13-156. LXVI 3-117. CCXXXVI 3- 134. Marcus Aurelius III 19-120. Ausonius General-177. Ep.: XII 40-139. Lud.: 104-140. Avitus-178. Boethius-179. Capitolinus, lulius Historia Augusta V 9, 10-145. VIH S, 2-153. Charlemagne-172n. 52; 180-181. Claudianus Mamertius General-178. De Statu Animae: I 6 (p. 42, 11)-120. I 6 (p. 42, 12)-120. I 9 (p. 49,18)-123. I 19 (p. 69, 1)-147. 1 19 (p. 69, 9)-148. Epilogus (p. 192, 22)-115. Columella VIII 5, 14-140. 1X1,7-141. 1X3,4- 130. Index Locorum 209 Commodian General-176. A.: 59-148. 487-140. 543-146. 736- 155. 787-152. Instr.: 117, 6-155. 124,7-136. 131, 1-121. I 33, 8-134. I 37, 10-140. H 15,13-136. 1119,2-155. Corippus General-172n. 52; 179. lo.: I 103-132. IV 626-149. VHI 207-128. lust.: 1181-122. Cyprian-178. Ennodius Geneial-178. 21,20-127. 83,19-135. 99, 3-145. 100, 14-135. 106, 26-119. 107, 4-119. 109, 4^128. 113, 5-130. 129, 21-119. 197, 3-123. 206, 36-111. 209, 32-131. 211, 6-135. 229, 3-152. 232, 20-116. 234, 21-121. 246, 1-112. 262, 23-111. 270, 17-128. 275, 24-132. 280, 23-150. 283, 20-152. 291, 22-132. 309, 2-152. Festus 219-132. Formulae Aiidegavenses-169; 173n. 57; 176. Frontmus-179. Gellius 9,2,5-128. 18,7,2-112. 19,8,6-151. Gregory General-161ii. 11; 170; 172n. 52; 179. De Cuisu Stellarum 16 (p. 863, 5)-124. Glor. Conf.: 5 (p. 751, 14)-l-t8. 8 (p. 753, 25)-128. 20 (p. 759, 12)-146. 20 (p. 760, 4)-120. 38 (p. 771, 25)-153. 49 (p. 777, 20)-121. 58 (p. 781, 25)-146. 77 (p. 794, 13)-151. 96 (p. 809, 20)-123. 108 (p. 818, 17)-124. Glor. Mart.: 9 (p. 494, 17)-149: 171. 77 (p. 540, 15)-140. 105 (p. 560, 13)-144. 105 (p. 560, 20)-145. 105 (p. 560, 3l)- 121. 105 (p. 561, 6)-121. Est. Fr.: H Prologus (p. 58, 2S)-151. n 3 (p. 65, 1)-156. n 9 (p. 74, 15)- 114; 171. II 13 (p. 80, 23)-146. Ill Prologus (p. 108, 13)-146. Ill 31 (p. 135, 123)-132. IV 11 (p. 148, 2)-156. IV 12 (p. 149, 21)-146. V 7 (p. 199, 1)- 146. V 39 (p. 232, 8)-149. V 43 (p. 234, 26)-115. V 49 (p. 240, 6)-146. VI 36 (p. 276, 13)-149. VI 40 (p. 280, 1)-119. VI 40 (p. 280, 7)-121. VI 40 (p. 280, 15)-115. Vn 15 (p. 300, 3)-156. VUI 13 (p. 333, 4)-183. VIH 31 (p. 347, 20) -149. Vin 36 (p. 351, 11)-156. IX 2 (p. 359, 22)-128. DC 38 (p. 392, 21)- 150. IX 49 (p. 240, 6)-146. X 1 (p. 407, 26)-119. X 2 (p. 410, 4)-150. X 3 (p. 411, 13) 135. X 16 (p. 427, 13)-116. X 19 (p. 43S, 10)-146. Mirac. B. Andr. Apost.: 1 (p. 828, 16)- 152; 171. 1 (p. 828, 22)-146. 4 (p. 829, 17)-148. 22 (p. 838, 35)-128. 23 (p. 839, 29)-122. 24 (p. 840, 18)-132. 24 (p. 840, 40)-126. 28 (p. 842, 39)-95; 151. Passio Sept. Dorm.. 7 (p. 850, 23)- 94; 148. Virt. S. lul.: 31 (p. 577, 6)-146. 33 (p. 578, 8)-lll. Virt. S. Mart.: I 22 (p. 600, 8)-119. I 23 (p. 600, 15)-145. I 28 (p. 602, 3)- 121. 1 34 (p. 604, 37)-147. 1 36 (p. 605. 25)-119. II 2 (p. 610, 9)-146. H 16 (p. 614, 12)-146. II 16 (p. 614, 21)-150. n 32 (p. 620, 29)-lll. n 40 (p. 624, 3)- 112. II 55 (p. 628, 3)-156. IV 29 (p. 656, 25)-146. Vit. Patr.: VI (7) (p. 685, 21)-119. VII (4) (p. 689, 27)-124. VHI (5) (p. 696, 7)-152. Vm (10) (p. 700, 5)-119. DC (p. 700, 2)-146. XX (4) (p. 743, 29)- 141. Hegemonius-179. Hermerus, Mulom, Chironis General-179. 38-130. 191-127. 352- 129; 136. 451-136. 567-135. 984-132. Heading in Hermerus, Mulom, Chironis 256-138. 210 Index Locorum Hilarius-177. Hyginus 167-134. Hyginus Gromaticus-179, Ildefonsus Toletanus-180. Itala-143; 167ii. 38; 179. Jerome General-170; 178. C. Pelag.: II 14-152. Ill 8-152. C. Vig.: 5-114. Ep. 22, 10-133. In Ephes. pr.-135. In Gal.: Ill ad 5, 13-113. In Naum ad 2, 1-127. Jordan-179. Juvenal V 67-121; 163n. 14 and n. 16. Lactantius-177. Lampridius, Aelius Historia Augusta XVI 7, 5-151. Lucan General-163n. 14; 163. I 126-114. II 682-1211. 11; 113. IV 201-128. VIII 170-117. VIII 645-136. IX 563-116. Lucifer of CagHari-177. Lucilius (?) Aetna 220-131. 231-153. Martial General-163n. 14 and n. 16. Ill 30, 2-110. Ill 32, 1-117. Ill 95, 3-112. V 58, 2-110. VI 88-117. IX 92, 1-135. Paterculus, Velleius-161n. 10. Paulus, Ed. 23, 2, 10-87ii. 2; 154. Persius III 67-112; 163n. 14 and n. 16. Petronius 8, 2-126. 33-139. 44, 2-157. 55, 3- 110. 67, 2-110. 71-141. 76, 11-127. Pliny the Elder, N. H. XXXVII (38) 119-129. Pliny the Younger-161n. 10. Priscian KeU II, p. 421, 20-127. II p. 423, 25- 127. PriscilUan-172; 173ii. 56; 177. Prudentius-177. Quintilian, Inst. General-161n. 10. IV S, 1-132; 172. IV 5, 26-128. Pseudo-Quintilian-176. Rutilius Lupus II 6-111. Scaurus, Terentius, in Varro, Reliquorum de Grammatica Librorum Fragm., p. 223- 133. Seneca the Elder Controversiae II 5, 14-127. X 2, 1- 137. Suasoriae 1 10—129; 171. Seneca the Younger General-xviii n. 7. DeClem.: 113,3-127. DeBen.: II 29, 1-122. De Paupertate 4, 10-134. Dial.: I 1, 1-115. I 2,5-156. I 4, 4^134; 172. 15,7-151. VII 25, 1-115. Ep. Mor.: I 4, 8-134. VI 6, 15-116- XrX 7, 7-135. XX 4, 21-129. Nat. Quaest.: II 22, 3-156. II 29-156. II 53, 3-147. Ill Praet. 9-150. Ill 10, 2-156. Ill 20, 2-88, 157. V 18, 12-114; 171ii. 51. VII 25, 1-153. Tragedies : Agam. 414-111; 159. Here. Fur. 1299- 122. Here. Oet. 365-147. Oed. 211-79; 140. Thy. 772-147. Tro. 811-139. Siculus Flaccus-179. Silius Italicus II 348-122. IV 71-150. XIII 446- 31; 123. Silvia (see "Aetheria") Statins Silv. IV 1, 25-142. Theb.: I 188-118. IV 333-117. V 734^126. XII 260-121. XII 805-118. Suetonius-161n. 10. Sulpicius Severus General-178. Dial. 1 2, 2-155. 1111,3-140. Tacitus General-160n. 5; 161n. 10. Index Locorum 211 Tertullian General-143; 176. Ad Mart.: 2-140. 6(end)-148. Ad Nationes: 1 1-135. Ad Uxorem I 8-123. Adv. Hermog.: 27-144. 45-137. Adv. Praxean 2 (end)-130. ApoL: 6 (begirming)-140. 13 (begin- ning) 140, 29 (beginning) 154. Bapt.: 15 (beginning)-155. Car. Chr.: 3 (p. 894)-110. 18-147. Cult. Fem.: 12-145. Idol.: 24(end)-141. Praescr.: 8 (beginning)-134. 22 (end) -133. Pud.: 13(end)-139. Virg. Vel.: 6 (beginning)-141. 9- 115; 171. Tiberianus II 18-137. Trebellius Pollio Historia Augusta XXII 5, 3-154. Valerius Flaccus VII 120-117. Valerius Maximus 5, 7 ext. 1-154. Heading in Valerius Maximus 8, 1-138. Vegetius General-180. IV 1, 13-152. IV 3, 9-115; 171n. 51. V, 15, 1-145. Varus to Fronto 116, 12-157. Victor Vitensis-172n. 52; 178. Vopiscus Historia Augusta XXVI 26,4^-131. XXIX 8, 9-130. Vulgate-143; 179. Index of Unknown Writers Cited from Modern Collections Bruns, C. G. and Gradenwitz, O., Fontes luris Romaui Antiqui S. C. Vellaeanum-l31. Buecheler, F., Carmina Latina Epi- graphica 73,1-99. 222,7-116. 369,1-123. 4S0a, 9-149. 457, 2-121. 565, 5-154. 596, 5-153; 171. 932, 2-129. 984, 2-144. 1000,2-146. 1136,1-116. 1381,13-147. 1489-122. 1839-122. Corpus Inscriptionum Latinarum II 4426-122. II 6130-136. VI 6435- 147. VI 9693-116. VIII 218-149. IX 1527-99. IX 5041-144. X 4183-116. X 5958-153. XI 4311-121. XII 825- 154. Huebner, E., Inscriptiones Hispaniae Christianae 219-112. 255-132. Pirson, J., La langue des inscriptions latines de la Gaule -145, 146, 147, 180. Pirson, J., Pamphlets bas latins du Vile Si6cle -180. Ribbeck, O., Scaenicae Romanorum Poesis Fragmenta II p. 122, 62-40. -^^[: .-^M5Vl..l,^ :^^ ■-avi;;,:-^/:*^