STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. THE SUBJUNCTIVE IN RESTRICTIVE QUI AND QUOD CLAUSES By Jefferson Elmobe We have first to note the use of the simple relative in the sense of an indefinite or generalizing one. This idiom, familiar in Greek, has been somewhat neglected in Latin. It is well illustrated in the lan- guage of law, of which it is, indeed, a characteristic. If one examines the texts contained in Bruns's Forties, he finds that quicumque and quisquis appear, comparatively speaking, but rarely, their places being taken by the forms of qui. This is shown in Lex agr. 71: "quam pecuniam qui agrum publicum emit .... populo dare debet," 1 and in Lex Cornelia i. 33: "quam in quisque decuriam lectus erit, is in ea curia viator esto." 2 In Lex Ursonensis xciii. 20 : " quicumque duo- vir creatus erit quive praefectus ab duoviro .... relictus erit," 3 we have qui repeating a preceding quicumque, showing the equivalence of the two words. So with quod in Lex Acilia 58: "ab eo quod quisque petet .... id praetor .... iudices aestumare iubeto." 4 In the literary sources this usage is noted by the grammarians 6 for different periods, but details showing its precise range and the conditions under which it appears have not, as it seems, been worked out. A somewhat special form of the idiom is the use of the pronoun with a dependent genitive. Cases of this occur in literature, but, like the construction in general, it receives its best illustration in legal texts where clearness and precision of language are the special aim. I will give a few examples with qui, the first showing that the con- struction was not unknown in Oscan : Leges Oscae Tabulae Bantinae 23, "alio famelo in ei(tuo) sivom, paei eizeis fust, .... tovtico estud," which in the Latin translation is, "cetera familia, et pecunia tota, quae eius erit .... publica esto"; 6 Lex Acilia 27, "ea nomina quei petiverit .... quei eorum volet, .... describendi is praetor potestatem facito"; 7 Lex Falcidia 1-2, "qui cives Romani 1 Bruns, p. 84. « Ibid., p. 68. • Brans, p. 52. » Ibid., p. 90. » Cf. Kilhner, 849, 12. ' Ibid., p. 64. > Ibid., p. 130. [Classical Philology XII, July, 1917] 253 254 Jefferson Elmobe sunt, qui eorum .... testamentum facere volet . . . . ius potes- tasque esto"; 1 Lex Malacitana Ixiv. 30, "omnes .... qui eorum liberati non sunt, erunt . . . . obligatisunto"; ibid.,lx. 41,"perquem eorum de quibus iivirorum quaestorumve comitiis suffragrium ferri oportebit, steterit quominus recte caveatur, eius qui comitia habebit, ne rationem habeto." 2 To these may be added some examples with quod: Lex Acilia 67, "quod eius is reus non solverit, ab iis prae- dibus .... pecunia exigatur"; 8 Lex agr. 5, "ager publicus, . . . . quod eius extra urbam Romam est, quod eius in urbe oppido vico est, quod eius iiivir dedet adsignavit .... publicus esto "f Lex agr. 26. "pro eo pecore, quod eius in callibus vieisve publicis pastum impulsum itineris causa erit nequid populo neive publicano dare debeto"; 6 Lex agr. 39, "id sententia pronuntiato quod eius .... maxsume verum esse comperrit"; 6 Lex agr. 95, "quodque in eo agro vini oleive net " ; 7 Lex Antonia i. 17 : " quodque earum rerum ieis consulibus iei habuerunt .... habeant"; Edicta praetorum x. 6, "in integrum restituam quod eius per leges .... licebit"; 8 ibid., xliii. 13 c, "si de ea re satis datum est, quod eius cautum sit ... . vim fieri veto." 9 In these examples it is to be noted that the genitive represents a whole (a thing, a group of things or persons) of which the relative expresses a part. Considered in their relation to the sentence it is the function of these clauses to limit the application of the predicate to a certain section of the subject. Thus in the passage from Lex agr. 5 the legal provision which the predicate embodies does not relate to the whole of the ager publicus, but only to such parts as are marked off by the quod-eius clauses. This limiting function, though assisted by the presence of the genitive, depends ultimately on the relative and the fact that it is used in this indefinite or generalizing sense. The clauses thus form a special category both in form and in function. To this category, in my view, are to be referred the clauses which are usually called restrictive. In some of these the relative still appears with a dependent genitive as in Cic. Fam. v. 8. 5, and Alt. xi. 12. 4, "tu tamen velim ne intermittas, quod eius facere poteris, 1 Brans, p. 110. « Ibid., p. 75. ' Ibid., p. 88 ' Ibid., p. 152. ' Ibid., p. 78. s Ibid., p. 217. « Ibid., p. 69. ■ Ibid., p. 81. • Ibid., p. 232. Subjunctive in "Qui" and "Quod" Clauses 255 scribere ad me." So we have in Lex Ursonensis lxxx, "rei rationen decurionibus reddito .... quod eius fieri poterit"; 1 and in Edictum Augusti 25, "quod eius sine dolo malo fiat." 2 It is thus practically- certain that this genitive (whether expressed or only felt) was originally a part of all the restrictive clauses. It follows from this that the relative itself must in all these cases have had a generalizing force, since otherwise it would not naturally admit of a dependent genitive. Moreover, the function of the two sets of clauses is virtu- ally the same. The idea of restriction is merely the limitation of the predicate of the sentence to the range of the restrictive clause; its affirmation is valid only within this range. This is the purpose of quod possum, potero, potui, quod in me est, fuit, quod ad me attinet, quod sciam, quod audierim, and the like; it is also the purpose of the quod-eius clauses in such sentences as "ager publicus, quod extra urbe Romam est ... . privatus est," 3 and in "integrum restituam quod eius per leges .... licebit." 4 Nor is there any essential difference (except the mode) between the relative clauses in "omnium oratorum quos quidem ego cognoverim acutissimum" 6 and "omnes .... qui eorum liberati sunt .... obligati sunto." 6 The relation I am here pointing out has been obscured by the fact that the stereo- typed phrases have for the most part lost the genitive of the personal pronoun, but it also makes clear that, considered with respect to their function in the sentence, the restrictive clauses constitute a much larger class than the grammarians have given us to understand. We are now in a position to deal with the subjunctive which with one exception 7 occurs in all the stereotyped forms. If the view is correct that the introductory relative is indefinite or generalizing, it follows that all these clauses are conditional, and that all of them have the possibility of the subjunctive. This fact at once accounts for the subjunctive in a large number of restrictive clauses, especially those relating to the future. I will give some examples. An excel- lent one is Cic. Fam. xvi. 4. 2, "sumptu ne parcas ulla in re quod ad 1 Ibid., p. 129; cf. Lex Ursonensis lxx and lxxi, p. 127. * Ibid., p. 250. » Lex agr. 5. « Edicta praet. x. 6. * Cic. Brut. 48. 180. • Lex Mun. Mai. lxxiv. 30 (Bruns, p. 152). ' Quod ad me attinet; Hale's view that the indicative only is found with certain predicate is erroneous. 256 Jeffekson Elmore valetudinem opus sit," where quod opus sit = si quid opus sit. To be understood in the same way is Lex agr. 2, " qui .... agrum .... sumpsit .... quod non modus maior siet quam quantum unum hominem .... ex lege .... sumere .... licuit." 1 Here belong the quod-fiat clauses, as in Cic. Att. xiv. 16. 4, "et ipse, quod commodo tuo fiat, cum eo loquare"; 2 and in Lex Acil. 69, "quoi pecuniam ex hac lege, quod sine pequlatu fiat, praetor .... darei iusserit .... dato"; 3 and Lex Antonia i. 10, "itaque ieis omnibus sueis legibus .... utei liceto, quod advorsus hanc legem non fiat." 4 Similar to these are the two following examples: Cic. Fam. i. 1. 3, "decrevit ut regem reducas, quod commodo rei publicae facere possis," and Cic. Fam. iii. 5. 4, " sed quod commodo tuo fieri possit .... magni interest ut te videam." A passage somewhat out of the usual order is Lex Ursonensis lxxxii. 38, " Quique iis rebus fructue erit, quod se emisse dicat, is in iuga singula .... HS C . . . . dare damnas esto." 6 One of the best illustrations is Cato agr. 148. 1, "quod neque aceat neque muceat, id dabitur." These quotations with varying predi- cates illustrate the working of the less vivid condition in restrictive clauses. We may now go a step farther and apply this explanation to quod sciam* in sentences relating to the future. As shown by the development of the other clauses quod sciam represents an original quod eius sciam, quod being not a cognate accusative, but a so-called outer object in a kind of limited apposition with the subject, object, or general idea of the sentence. Its use in a future relation is illus- trated in Plaut. Capt. 265, "non ero falsiloquos [i.e., nihil falsum dicam] quod sciam." 7 This is only another way of saying, "If I 1 Cf. Lex agr. 13; Collegium aquae 6 (Bruns, p. 394). 2 Cf. Cic. Fam. iv. 2. 4; xiii. 2. 1; xiii. 23. 2; xiv. 5. 1. Att. xiii. 11. 2. ' Bruns, p. 70. 1 Ibid., p. 93; cf. Lex Acil. 45; Lex Mun. Salpensana 27; Edictum Augusti 25; Lex Luci Spolatini 8; Lex arae Augusti 14; Paul. Dig. xi. 5. 2. 1, quoted (and wrongly interpreted) by Kalb, Wegweiser, p. 124. s Bruns, p. 129. « Quod scio seems not to occur. Expressions of similar meaning with the indicative appear sporadically. Thus we find quae (=as far as) mihi conscia sum (Ter. Eun. 457) quod jmto (Cic. Att. 15. 2. 2.), quod sentiunt (Lex Furfensia v. 10), and quod video (Juvenal vi. 395), but I have found no case of quod scio. That it was not current speech seems further evidenced by the French que je sache. 7 Cf. Bennett, Syntax of Early Latin, p. 295. Subjunctive in "Qui" and "Quod" Clauses 257 should know anything of the matter to be false, I will not utter it." The same idea of a future contingency is in the restrictive qui clause used by Cicero in Cat. iv. 16, "servus est nemo, qui modo tolerabili condicione sit servitutis, qui non audaciam civium perhorrescat." In fact, it seems to be true that in all restrictive clauses relating to the future the subjunctive where it appears is sufficiently accounted for by the conditional character of the clause. We have still to explain the presence of restrictive clauses with the subjunctive in sentences relating to the present or the past. How does it come that quod sciam appears in these contexts with apparently a purely indicative meaning? (Cf. Cic. Att. xvi. 2. 4, sed non venerat, quod sciam.) One way of approach is to suppose that we are dealing here with a different kind of subjunctive. I must confess that for a long time this appealed to me strongly, and I made every effort to find some original use of the subjunctive which could be adjusted to this particular situation. I was especially hopeful of the potential subjunctive as expressing possibility, but I found that my attempts to apply it to these clauses resulted in a funda- mental contradiction. In fact, it might have been seen from the beginning that it was improbable a priori that two different uses of the subjunctive should be represented in the same construction. If this be the case, there remains only the hypothesis that quod sciam relating to the present and past is the outcome of its use in the future. Fortunately, this hypothesis, reasonable in itself, can be supported by several considerations. It implies (1) that quod sciam originated (and perhaps became stereotyped) in future contexts in sentences of the type of "omnem rem tibi narrabo, quod sciam," (2) that later it became transferred to other ranges of time, (3) that in adapting itself to the new environment it virtually lost its original modal significance. The first point offers no difficulty whatever; it is, indeed, what we might expect, seeing that the subjunctive here is perfectly appropriate and normal. The second step would be facilitated by the general requirements of everyday speech and in particular by the absence of quod scio. The mode itself, if felt merely as giving a tinge of uncertainty to the clause, would not be altogether out of place. Still, it must be recognized that in time the clauses became indicative in meaning, but this modal transformation 258 Jefferson Elmore is by no means strange. "Mood," as a recent student of the sub- ject points out, "varies with its surroundings; the meaning suggested by the context predominates Of all the inflectional variations of the verb it is the least definite in character and yields most readily to outside pressure." 1 In Latin one has only to recall the relative clause of characteristic, which, though retaining the subjunctive form, has come to be a mere statement of fact. In clauses with -priusquam and the like, where the action is anticipated and prepared for, Hale calls attention to the frequent use of the subjunctive to represent a real event. After Cicero's time and as early as Livy the distinction between the subjunctive and indicative in this construc- tion broke down. 2 Hale has also called attention to the use of the subjunctive in the second person singular in general statements of fact. Here, so far as meaning is concerned, a subjunctive has passed into an indicative, the change affecting both dependent and inde- pendent clauses. 3 A similar process is illustrated in the change which has taken place in the velim and vellem forms. 4 These forms, through being subjected to new influences, have in certain cases become essen- tially indicative in meaning. There seems then to be no reason why a similar change under the circumstances should not have taken place in quod sciam, and also in other types of the restrictive clause when placed under like conditions. Stanford University 1 McWhorter, Proc. Amer. Phil. Association, Vol. XLV, p. xxv. » Lai. Gram., 507, 4, n. 1. s Class. Phil., I, 21 f. • Morris, "The Subjunctive in Independent Sentences in Plautus," Amer. Jour. Phil., XVIII, 133, 275, 285, 383. See also the present writer's Syntax of Certain Latin Verbs of Desire, p. 76.