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 byJSTOR. 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.istor.org/participate-istor/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. CLAUSES OF WILLED RESULT By Frank H. Fowler It is the purpose of the present paper to point out the interrela- tions of certain ut and ne clauses existing in Latin. Incidentally the paper is a protest against two tendencies seen in recent syntactical investigation, both productive of confusion and error: (1) the ten- dency in the setting up of categories under which clauses may be classified to neglect ideas expressed other than the one taken as the distinguishing mark of the category,' and (2) the tendency to seek for a separate origin in parataxis for every category set up.^ A sentence such as iVe deterius huic sit expresses the will or the wish of the speaker in opposition to some act or state. Such a sentence may be uttered without the consciousness of any relation between it and a preceding sentence, notwithstanding the fact that in the psychological process back of the expressions some relation necessarily existed. When, however, the speaker does become con- scious of a relation and expresses it, it will be some particular relation and not relation in general. Our initial task, then, is the discovery of the relation or relations which might come to be recognized as existing between a ne sentence and its antecedent, which relation then might come to be actually expressed when the ne sentence came to be felt as subordinate. Two cases may be distinguished. (1) The verb of the accompanying sentence might be one capable of taking a ne sentence with its volitive or optative meaning as an "inner object" or "inner subject." With clauses resulting from the recognition of such a relation we are not here concerned. (2) If the verb of the accompanying sentence was one not capable of taking a ne sentence as an "inner object" or "inner subject," no relation could be recog- nized unless there existed between the two acts or situations a "causal nexus," that is, unless the act or situation of the ne sentence were > Cf. Morris, Principles and Methods, pp. 206 f. 2 Cf. Frank, Classical Philology, III, 13. In the opinion of the present writer, the evil result of both of these tendencies is to be seen in Professor Bennett's failure to properly relate the " stipulative " to other tti and ne clauses. This paper is especially concerned with the determination of that relationship. [CiiAssicAL Philologt XV, January, 1920] 46 Clauses of Willed Result 47 seen to be the result of the act or situation of the accompanying sentence.' The ne sentence became subordinate when the expression of this relation came to be felt as a function of the sentence. In mode- rare animo: ne sis cupidus of Mil. 1215 it is a question whether the act of moderare animo tends to bring about the state of ne sis cupidus or not. If we decide that it does not, we also decide that ne sis cupidus is not a subordinate clause of any kind. But if one say, "I want him to be cared for," "I don't want him to be worse off," it is seen at once that "not being worse off" is the result of the "caring for " ; and it is, as stated, a result willed. The ne clause of the follow- ing passage is property classified as a ne clause of willed result. Capt. 738: Atqui hunc me velle dicite ita curarier Nequi deterius huic sit quam quoi pessumest. Unfortunately the terms "result" and "consecutive" have been restricted very largely to use as designations of a certain class of clauses, the so-called clauses of "actual"^ result. But it is possible for the consecutive relation to exist between an antecedent verb and another verb having any modal meaning.' This is evidenced in Eng- lish by the possible combination of "so that" with any of the modal auxiliaries. The existence of ne result clauses has been asserted by Brix (note to Mil. 149), Morris (Introduction to Captivi and Trinummus), Sonnenschein {Unity, p. 37), and others. The contention of this paper, however, goes much farther. Not only are there ne clauses which, since the consecutive relational meaning is predominant, may be classed as result clauses,* but the consecutive relational meaning was of necessity the first to be established, and that meaning more or less clearly maintains itself in all adverbial ne clauses and in sub- stantive clauses derived from them and gives unity to their history. A study of ne clauses along lines thus suggested should serve, among other things, to place the stipulative clauses in their proper per- spective. ' Cf. Methner, Neue JahrbUcher fur POdagogik (1909), pp. 185 f. 2 By the use of the quotation marks with this word here and elsewhere in the paper I indicate my dissent from the common opinion as to the modal meaning of the verb in these clauses. 5 Contrast Bennett (Critique, p. 9), who criticizes Elmer's (essentially correct) statement that "every purpose clause is a result clause as well." * Cf. Morris, Principles and Methods, p. 213. 48 Fbank H. Fowlek First should be mentioned other clauses of willed result like Capt. 738, quoted above, clauses in which the consecutive relational idea is especially prominent. The clearest cases are those in which ita or something similar appears in the antecedent clause. Bacch. 224: Veniat quando volt atque ita ne mihi sit morae. In such sentences other relational ideas come in very easily, and this particular example will be quoted again under a different heading (p. 53). Bacch. 224 together with Cure. 29, Merc. 146, Trin. 343, Capt. 267 (ut ne), Merc. 960 (ut ne), are classed by Sonnenschein {Unity, p. 37) as consecutive. Morris in the Introduction to his edition of the Captivi and Trinummus in addition to Capt. 738 and 267 so classes Most. 390, Mil. 149, and Most. 1050, the last two having ut ne. An example with ita and ut ne is the following: Phor. 975: Hisce ego illam dictis ita tibi incensam dabo, Ut ne restinguas, lacrimis si exstillaveris. This is so clearly a result clause that Bennett {Syntax of Early Latin, 1, 297) so classifies it and adds, "if ut ne is what Terence wrote, it is unaccountable and unexampled." But the result was willed and therefore necessarily ne was used just as in Ita corrigas librum, ne mihi noceat (Caecina apud Cic. Fam. vi. 7. 6).^ In some cases we may be sure that the consecutive relational idea is the only relational idea expressed, or at least that it is predominant, even though the main clause contains no such word as ita. Bacch. 701: Emungam hercle hominem probe hodie, ne id nequiquam dixerit. Compare Eun. 278 and Phor. 885. In interpreting the positive ut clause we have the difficulty that nothing in the clause itself indicates the modal meaning of the verb. However, the modal meaning may be indicated with some degree of certainty by the character of the antecedent sentence. So in Aul. 591, Sin dormitet, ita dormitet, servom se esse ut cogitet, the clause is one of result and the verb is volitive. Compare Asin. 794 and Pseud. 929, and compare this last with Phor. 975 quoted above.^ • Cf. Sohmalz, Lat. Gram., p. 513. ' A positive clause of willed result in Umbrian is seen in vi o 20 : Eo iso oatendu, pusi pir pureto cehefi dia, " Let him set them out in such a manner that he cause one fire to be lighted from the other." Cf . Buck, Gram, of Oscan and Umbrian, p. 216. Clauses of Willed Result 49 To the relational idea of willed result other relational ideas were easily added and made predominant; from the ne clause of willed result were easily developed other ne clauses. Especially easy was the development of the purpose clause, so called. An act or state seen as the undesirable result of another act will very commonly furnish the reason or motive for that other act. A ne clause express- ing a result willed against may well serve as a motive clause. Such an example as the following, then, may be called a ne clause of willed result as motive: Men. 852: Adserva istunc, mi pater, Nequo hinc abeat. The ease with which an expression of result may come to express also motive may be illustrated by the use of the Greek dicrre-infinitive phrase as in Plato Gorg. 479 C, Hav Troiomiv, wcrre Sikijj' /tj) hbbvai. Other examples may be found in Goodwin, Moods and Tenses, § 587, 3. The ut non (and ut) substantive clauses after such verbs as facio are without doubt substantivized clauses of "actual" result.^ For example, in And. 17, Faduntne intellegendo ut nil intellegantf the ut clause, besides giving the result of the action, expresses also the object effected by that action. In the same way the ne clause of willed result was substantivized after verbs meaning "to produce a result," "to cause." The substantive clause after such verbs could not be primary, could not arise from an original parataxis, for the reason that an expression of an act or state willed against is in itseK incapable of becoming the "inner object" of such a verb as facio. But, on the other hand, a clause such as ne metuam employed to express adverb- ially the willed result of another verb would, in case that verb were fac, be seen to express also the object effected. The following, then, is an example of a substantivized clause of willed result: Most. 1145: Fac ego ne metuam. Other examples are Poen. 909, Hec. 839, and Most. 423 (ut ne). Sentences such as Most. 390, Satin habes, si ego advenientem ita patrem faciam tuom non modo ne intro eat, verum etiam ut fugiat longe ah aedibusf in which the verb has a noun object beside the ne or ut clause, show clearly the close relationship existing between • Cf. Hale-Buck, Grammar, 521, 3 a, and Bennett, Syntax of Early Latin, I, 299. 50 Frank H. Fowler the adverbial and substantive clauses. The clauses in the example quoted are clearly consecutive;' if they are not also substantive, the change of patrem to pater would make them so. In the case of the ut substantive clause after forms of facio there should be no question concerning the volitive character of the clause when the antecedent sentence has fac, facite, or other clearly volitive expression. Capt. 337: Fac is homo ut redimatur. From the nature of the case we should hardly expect to find substantive clauses of willed result dependent on verbs meaning "to be caused" or "to happen." However, the clause does occur with potest fieri. Add. 626: Fieri potis est ut ne qua exeat. Compare And. 699 (with ut ne) and Poen. 1253 (with ne). The classification of these clauses as substantive clauses of willed result recognizes the substantive character which Bennett {Syntax of Early Latin, I, 238) sees in them, the result idea which Fairclough sees (so also Freeman and Sloman) when he mistakenly says (note to And. 699) that id ne equals ut non, and the volitive character back of Ashmore's interpretation of the Terence passages as clauses of purpose. And further in Cic. Verr. v. 5 we have an example of factam (esse) ne. The absence of ut from many of the positive clauses after /ac, etc., might seem to militate against the theory that the positive sub- stantive clause was secondary. The ut was not original in primary substantive clauses, while in positive adverbial clauses the ut was always present. If the ut adverbial clause became substantivized we might expect the ut to be kept. The explanation is that these secondary substantive clauses followed the analogy of the primary clauses omitting the ut at will. On the other hand the presence of a considerable number of secondary substantive clauses with an ut carried over from their use with adverbial function was responsible 1 So Morris, Introduction to Captivi and Trinummus, sec. 36. Bennett, Syntax of Early Latin, I, 228, denies the consecutive relation. Sonnenschein ( Unity, p. 46) compares Verg. Eel. iii. 51 , Effldam posthac ne quemquam voce lacessas with Cic. de Fin. ii. 5. 15, Rerum obscuritas faeit ut non intelUgatur oratio. But Sonnenschein's whole theory of subjunctive meanings keeps him from making a sharp distinction between result with volitive modal meaning and result with other modal meanings. Clauses of Willed Result 51 for the introduction of the ut into primary substantive clauses. A leveling in this respect took place between the two classes of sub- stantive clauses. This seems to me a better explanation of the presence of the ut than that of Professor Hale that the conjunction came in from the purpose clause.^ In the case of the clause of "actual" result it may become the chief function of the clause to characterize the action of the ante- cedent clause as, for example, in Suet. Aitg. 56, Numquam filios suos populo commendavit ut non adiceret ("without adding").^ For examples like Cic. Div. in Caec. xiii. 44, Cuiu^s ego ingenium ita laudo, ut non pertimescam, ita probo ut me ab eo delectari fadlius quum decipi putem posse, Bennett (Trans. Amer. Philol. Assn., XXXI, 248) gives the all too easy explanation "ut non for ne," disregard- ing the fact that the subjunctive is clearly not volitive and that the following ut clause is clearly consecutive. The ne clause sometimes appears as a ne clause of willed result characterizing the action or state of the antecedent clause. Poen. 178: Locum sibi velle liberum praeberier, Ubi nequam faciat clam, nequis sit arbiter. Compare Pseud. 321, Aul. 358, and Asin. 718. The idiomatic English equivalent of such clauses is the phrase composed of without and a verbal. This phrase is used to characterize an action. "Without flinching" in the sentence, "He does this without flinching," answers the question, "How?" But the phrase is modally colorless and in so far fails to translate the ne clauses of which examples have just been given. Instructive is the use of the ne clause co-ordinated with a sine phrase in Merc. 145 and Most. 414 (m) and its use to explain hoc modo in True. 919. The positive clause of "actual" result also was used with charac- terizing function as in Pliny Epp. i. 20. 6, Ita eludit ut contendat. There are but few cases in which one can be sure of a volitive modal meaning in a result clause so used, but the following example may be > Cf. Hale-Buck, Grammar, p. 261, n. 2. 2 The Latin quin clause sometimes has this same characterizing force, as does also the English clause introduced by biU or but that as in "It never rains but it pours" ("without pouring"). 52 Frank H. Fowler classed as a positive clause of willed result characterizing the action or state of the antecedent clause: Persa 35: Facere amicum tibi me potis es sempiternum. Quem ad modum ? Ut mihi des nummos sescentos, quos pro capite illius pendam. "So be it you shall give" ("by giving"). A special development of the ne consecutive-characterizing clause is to be seen in those ne and nedum clauses commonly trans- lated "much less," of which no completely satisfactory explanation has heretofore been given.^ These clauses are parenthetical, and in this respect resemble ne and id clauses of motive giving the motive, not of the main verb, but of the saying. The following should be classed as a parenthetical characterizing clause of willed result: Amph. 330: Vix incedo inanis, ne ire posse cum onere existumes. " To such a degree is this true that you shall not think." No exam- ples with nedum occur in Plautus/ and Terence's one example is Haut. 454. The ut clause of "actual" result following a clause with ita was sometimes used to express a limitation as in Cic. Cat. iv. 7. 15, Qui ita concedunt, ut vobiscum de amore rei publicae certent ("who yield only to the extent of vying").' This is a shift in the meaning of ita fully as much as it is in the meaning of the ut clause. (Compare the shift in the meaning of the adverb tantum.) The clause of willed result was used in the same way,* and when so used may be called a limiting clause of willed result. Liv. xxii. 61. 5: Ita admissos esse, ne tamen iis senatus daretur. The following is a clear example of the restrictive use of the positive clause of willed result: Cic. Tusc. i. 45. 108: Id curent vivi, sed ita, ut intellegant nihil id ad mortuos pertinere. ' Cf., however, Walden, Harvard Studies, II, 103 ff. 2 The three examples given by Weise, Lex. Plaut., of ne =nedum do not belong here. Stolz and Schmalz, Lat. Gram., p. 559, note, gives a number of Lucretian examples of nedum clauses. ' In Cic. pro Scauro iv. 5, Qui tamen ipsi mortem, ita laudant, ui fugere'vitan vetent, Bennett (Trans. Amer. Phil. Assn., XXXI, 228) fails to see that no volitive force is present and so classes the example as one variety of his stipulative clauses, all of which, according to him, have the volitive subjunctive. * Cf. Draeger, Hist. Gr., II, 630, and Methner, Neue Jahrbucher fUr POdagogik (1909), p. 192. Clauses of Willed Result 53 In the case of a clause of "actual" result used to characterize the action or state of the antecedent clause, if the action of the result clause lies in the future, the manner ascribed to the antecedent action may be seen to be one stipulated. We have in Plautus at least two examples of negative non-volitive stipulative clauses: Bacch. 1184: Quem quidem ego ut non excruciem, alterum tantum auri non meream. Persa 786 is undoubtedly similar, though the principal clause is not expressed. In Men. 216 and Stick. 24 we have similar stipulative clauses, and the subjunctive without much doubt has the same modal meaning. The ne clause of willed result indicating the manner stipu- lated is to be seen in the following. Bacch. 224: Veniat quando volt atque ita ne mihi sit mora. This clause Bennett classes as stipulative and Sonnenschein (Unity, p. 36) as consecutive. There is the same difference of opinion regard- ing Merc. 960 (with ut ne). Both are right. The translation might be, "so that there shall be no delay," "without delay," or "with the understanding that there shall be no delay." Haitt. 783 : Ita tu istaee tua misceto, ne me admisceas. "Mix up your mixings so you mix me not withal." (Gildersleeve- Lodge, Latin Grammar, §552, 3). As an example of the correspond- ing positive clause may be given the following: Epid. 469: Tuas possidebit mulier faxo f arias, Atque ita profecto, ut earn ex hoc exonere agro. That the stipulative clauses are result clauses with a special function is indicated by the use of the Greek cjo-re-infinitive phrase in such a sentence as the following: Thuc. iii. 28, Hoiovvrai biioKoyiav TTpds Hax'ip'O; (bare 'A6r}vaioi.s i^etvai jSovXevcrai irepl twv M.VTiKrfva'uiiv. Other examples will be found in Goodwin, Moods and Tenses, §587, 2. From this study it appears that those ne clauses (and the corre- sponding ut clauses) which were not primarily substantive clauses originated as clauses of willed result, that to the consecutive relational meaning other relational meanings were added very much as in the case of the clause of "actual" result, and that therefore the several ne clauses, as the purpose clause and the stipulative, did not origi- nate separately from a parataxis but rather are to be thought of as phases of the clause of willed result. The Unitbbsitt op Arizona