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 CONSECUTIVE CLAUSES By J. J. SCHLIOHBB The current explanation of the Latin subjunctive in consecu- tive clauses holds that it is in its origin a potential, or mood of limited or modified assertion. 1 This explanation is unsatisfactory for several reasons. In the first place, it does not account in a convincing manner for the use of the mood in those consecutive clauses which express a fact, as well as in those which express a tendency or possibility. Elsewhere in the language the potential subjunctive represents things as likely to happen, as possible, as sure to happen under conceivable conditions, and the mood thus serves to distinguish things of this sort from those which actually do, did, or will happen, and which are expressed by the indicative. How could the subjunctive come, in consecutive clauses, to express both kinds of ideas, to the entire or the partial exclusion of the indicative ? This assumed extension of the powers of the potential is spoken of 2 as "an easy transition," and a somewhat detailed explanation of it is given by Professor Hale in one of his articles on "The Sequence of Tenses." 3 According to this view the consecutive clause was originally an independent expression of limited asser- tion, which in due time became a subordinate clause of limited assertion. And finally this subordinate clause of limited assertion became one of unlimited assertion. This last step in its develop- ment the writer explains as follows: "Next there intrudes into the idea conveyed by the construction, which does not in itself deal with the world of reality, an idea that squints at that world. If I say, 'He is such a man as never to lie,' I might as well have said, 'He is a man who never lies,' and might, indeed, very easily 1 Bennett Appendix to the Latin Grammar, pp. 200, 201, 216. Lane Latin Gram- mar, § 1818. Hale "The Sequence of Tenses," Am. Jour. Phil. VIII, p. 49, and "Oum-Construktionen," p. 120; Allen and Greenough Latin Grammar, §§ 534, 536. 2 Bennett Appendix, p. 202. *Am. Jour. Phil. VIII, p. 51. [Classical Philologtt II, January, 1907] 79 80 J. J. SOHLICHEK be quoted as having said that precise thing. In many Latin sen- tences, in fact, it is impossible to be sure whether limited or unlimited predication is meant. And so the thing felt and the thing said come to be confused, and the construction of the latter is used to express the idea of the former; or, in other words, in the Latin language the mood of characterizing predication limited becomes also, in relative sentences, the mood of characterizing predication unlimited." The weak point in this explanation is the absence of a good reason why this extension of the powers of the mood should take place only in consecutive clauses. Why, for instance, did not the independent sentence, "He would not lie," in the same manner crowd out the indicative, "He does not lie" ? Or, for that matter, why did not the indicative, in its turn, crowd out the subjunctive ? Apart from this, it is perfectly true, of course, that one may sometimes express himself about certain facts or circumstances with equal veracity either by the subjunctive or the indicative, and this is especially true of general conclusions, like the one in the illustration just quoted, which are drawn from a certain amount of evidence that the speaker has before him. The choice of mood will depend on whether the speaker is habitually out- spoken or cautious; it will depend on his own mood at the time, on his audience, on his relation to the matter under discussion, or on other conditions too numerous to mention. It is in such a case primarily not a question at all of the bare external facts. For it occurs rarely, if at all, that the speaker has an absolutely complete knowledge of these facts. We may say that a certain man never lies, but if we do so, we speak with disregard of the necessary incompleteness of our knowledge; we express the truth so far as we see it, but in a manner as if we saw it all. If we are at all cautious, we will say that the man "would never lie." There is a vital difference, then, between the indicative state- ment and the subjunctive, to the mind of the speaker and to the mind of the hearer, even though the external facts about which that statement is made are the same. And this is the really essen- tial point. For language is the expression of what is in the mind, not necessarily of what is outside, of the mental attitude toward The Subjunctive in Consecutive Clauses 81 or conception of the external world, not of this world itself. Now this difference of conception or attitude was defined and accentu- ated in Latin by a consistent difference in the form of the verb. There was just as much difference between the indicative and the subjunctive in Latin as there is between "does not lie" and "would not lie" in English. And as the subjunctive was in the language generally associated with qualified as opposed to unqual- ified assertion, it will be necessary to show a very special reason in the case of the consecutive clause, if we are to admit that the distinction was broken down here only. This will not be an easy thing to do. For it is certain that there is nothing in the nature of the consecutive relation itself that would make such a confusion natural or probable. The domination of the subjunctive is in fact a characteristic peculiar to the subordinate consecutive clause alone. For in those cases where the consecutive relation exists between independent, or rather paratactic, clauses, the dictinction between indicative and potential subjunctive is everywhere tenaciously observed. 1 There are, of course, among the subordinate consecutive clauses also many examples in which the force of the subjunctive is really that which goes by the name of potential. But if we insist on thrusting this fact into the foreground of the discussion, and thus bring ourselves to a sharp realization of the gap between these clauses and those in which the mood expresses a fact, and if we then try to bridge this chasm without further ado, we are merely looking at the problem from a difficult and unprofitable point of view, and are attacking it from a side which promises the least possible hope of success. This difficulty is, however, but one of those which the "po- tential" theory has to contend with. There is still another, which is even more formidable. Certain facts about the mood in consecutive clauses are made tolerably clear by the evidence. In general it is plain that the subjunctive has the upper hand. However, if the subordinate clause begins with a relative and the main clause is positive, a certain field is still left for the indica- 1 Compare, for example, Aul. 410: totus doleo atque oppido perii: ita me iste habuit senex gymnasium, with Capt. 464, 465: nam hercle ego huie diei, si liceat, oeulos eefodiam lubens : ita malignitate oneravit omnia mortalis mihi. 82 J. J. SOHLIOHEK tive, though as time goes on the subjunctive appears to be gain- ing ground here also. On the other hand, if the main clause is negative, then the subjunctive alone is used in the subordinate clause from the beginning. The most favorable environment for the consecutive subjunctive was then clearly a negative in the main clause. And the primary and original force of the mood itself, therefore, must necessarily be one which will fit these peculiar conditions — that is, one which will naturally go with a negative in the main clause. And here the subjunctive of limited assertion is entirely out of the question. For an assertion such as the potential makes, limited and qualified though it may be, is nevertheless an asser- tion. It is limited only because the speaker's power, or the evi- dence at his disposal, from which he draws his conclusion, is limited. And a statement of this sort, to which the speaker commits himself to the full extent of his knowledge or his power, is surely the direct opposite of a denial, or a negative. An illustration will perhaps make the point clearer. In the discussion by Professor Hale which has just been referred to, the development of the consecutive clause from the paratactic to the hypotactic stage is illustrated by these steps: "(a) You'd easily understand it; it is a very simple thing. (6) It is a very simple thing; you'd easily understand it. (c) It is a very simple thing so that you'd easily understand it." 1 That would of course be a sufficient explanation in this particular case. But this is not the sort of case that most loudly calls for explanation, for it is clearly not the earliest developed or the most typical form of the consecutive sentence. The earliest and most typical form would rather read: "It is not such a simple thing that you'd easily understand it." And if we reduce this to the paratactic stage by retracing the steps just taken, we arrive at an absurdity: "You'd easily understand it; it is not such a simple thing." Now, to be sure, an effort may be made to escape this conclusion by assuming that a sentence like nihil est quod dare possim is for practical purposes equivalent to nihil dare possim. To this we need only say that there is no more significance in it lAm. Jour. Phil. VIII, p. 50. The Subjunctive in Consecutive Clauses 83 than in the assumption that the English sentence, "There were two trees in the street which the men have cut down," is for practical purposes equivalent to, "The men have cut down the two trees in the street," or in the reduction of any complex sentence to a simple one. The two are equivalent "for practical purposes," it is true, but only in the sense that an ox is for prac- tical purposes equivalent to a horse. And besides, the inter- changeability of these two forms would not enable us to take a single step toward explaining why the consecutive clause with the negative antecedent was the earliest in point of development, or why it should exclude the indicative, while the one with the positive antecedent admits it. Still there is a grain of truth, after all, in the view that the main clause in these sentences is not of preponderant importance. When we examine the cases where a consecutive relation exists between two paratactic sentences — as, for example, "He will not (or would not) do it; he is such an honest man" — we find that the part which corresponds to the main clause of the hypotactic form almost invariably stands last. This clause, "he is such an honest man," is in fact, when we come to look at it, nothing more than an additional effort to impress the preceding statement, "he will not do it," more forcibly upon the mind of the hearer. Another method is employed, but the speaker's object is the same as in this previous assertion. The point which we need to keep in mind, then, is just this, that the second of these two assertions, which becomes the main clause of the fully developed consecutive sentence — in the present case, "he is such an honest man" — contains in reality nothing more than a reaffirmation of what is already expressed, less distinctly perhaps, and less forcibly, by the first assertion, "he would not do it." Now, in the earliest type of the consecutive sentence, as we saw, this originally supplementary or reinforcing assertion, which later became the main clause, contains a negative, or the equiva- lent of a negative. And if this negative did serve to express more distinctly and forcibly the same idea which was expressed by the other clause, surely that clause must then likewise have been negative in its force. So in the sentence, "It is not so cold that 84 J. J. SCHLICHEE you need an overcoat," the speaker's attitude toward the idea of the subordinate clause, "that you need an overcoat," is undoubt- edly a negative one: he refuses to admit that "you need an over- coat;" but this refusal, this negative attitude, is most clearly and distinctly expressed in the accompanying clause, by the word "not." Our next question must be to inquire what use of the subjunc- tive that can be which expresses this negative attitude ; for, as we know, in these negative forms, which were at the same time the earliest forms, we find no indicatives at all. And the answer must undoubtedly be that it is the same use of the subjunctive which expresses an unwillingness to accept or to vouch for a foreign idea in other cases, in indirect discourse, for instance, in the rejected reason after non quo or non quia, and in the subjunctive exclamations of surprise or indignation. Of these various construc- tions the most instructive and profitable for comparison, because the most primitive, is undoubtedly the indignant exclamation. 1 A very close parallel can easily be shown to exist between this excla- mation and the consecutive clause. For a considerable number of different types of this clause can readily be duplicated in para- taxis, as it were, with the exclamation standing in the same rela- tion to its paratactic companion, as the consecutive clause stands to its principal clause. 2 The type ita (tarn, tantus) ut (Hor. Sat. ii. 5. 18) : utne tegam spurco Damae latus? haud ita Troiae me gessi. The type factum ut (Plaut. Men. 683, 684): mihi tu ut dederis pallam et spinter? numquam factum reperies. 1 There is a wide difference in meaning between the exclamations which have the indicative and those which have the subjunctive. The former do not express a nega- tive or unwilling attitude, or reject another's word or thought, but rather reflect an inability on the part of the speaker to grasp the thought fully or readily. A full com- parison between the two has been made by the writer in an article, "The Moods of Indirect Quotation," Am. Jour. Phil. XXVI (1905), pp. 76-79. 2 A relation between the subjunctive exclamation and the consecutive clause, as well as many other clauses, has been suggested by A. Dittmar in his Studien z. lat. Moduslehre (1897). While in the first of these points his theory coincides with that here presented, the present writer does, of course, not wish to be understood as agree- ing with Dittmar's interpretations or as accepting his general position. Indeed, it is but fair to say that the views here presented had been formed in their main outlines, and in large part elaborated, before the writer became acquainted with Dittmar's book. The Subjunctive in Consecutive Clauses 85 The type quam ut (Ter. Phorm. 955, 956): hicine ut a nobis hoc tantum argenti auferat? tam aperte inridens? emori hercle satius est. The type non verisimile ut (Ter. Hec. 139, 140) : sese ilia abstinere ut potuerit ? non verisimile dicis ne verum arbitror. The type istuo [hoc, Mud) ut (Plaut. End. 225): Pe. utin impluvium induta fuerit? Ep. quid istuc tam mirabilest? The type is qui (Cic. Phil. vi. 3. 5) : huic denuntiationi ille pareat? ille se fluvio Kubicone et CC mili- bus circumscriptum esse patiatur? non is est Antonius. It will be well to inquire a little further, just what this sub- junctive of negative attitude involves. For it does really involve much more than the attitude of absolute refusal which is repre- sented by such types as non ita ut and nemo est qui. In an excla- mation of surprise, as in primitive expressions of emotion gen- erally, the extent of the application is necessarily but very vaguely defined. The exclamation, "He an honest man!" may mean, "Never!" but it may also mean," Yes, sometimes;" or, "Yes, but he is the only one !" or, "Yes, if it is to his advantage!" That is, in addition to the purely negative attitude, the absolute refusal to entertain the idea, the exclamation may even amount to a limited acceptance of it, provided that in the main the speaker's atti- tude is still negative. In relative clauses, for example, we should find the subjunctive used first and exclusively in cases like, "There is no one who wants to go;" then further in, "There is only one who wants to go," and, "He is the only one who wants to go," "There are but few who want to go," and, "They are the only ones, the first ones, the last ones, who want to go," "There are not many who want to go," and finally, "There are some, we admit, who want to go," or simply, "There are some who want to go." An instructive parallel of such an expansive use of the mood is found in the subjunctive of indirect discourse, where the same form is employed to express the absolute refusal to accept a foreign idea, on the one hand, and from that all the way down to cases where there is really no such refusal on the 86 J. J. SOHLICHEE speaker's part, but where he is merely assigning the idea to its proper author. But in addition to this growth of the construction by the natural expansion, as we may say, of the embryonic form, there is another, which must not be denied its due share in the develop- ment. After reaching the stage of subordination or hypotaxis, a construction acquires new rights. It may now, for instance, change from negative to positive, or vice versa, with perfect free- dom, even though originally the one, or the other, was the very basis and condition of its existence. So we may say, non hortor ut, although the clause with ut in the beginning presupposed exhortation itself, and not the absence of it. Just so an original nemo est qui might develop a positive form est qui, sunt qui, or even multi sunt qui, though the latter forms, with the subjunctive at least, when stating a fact, may have been originally impossible. Having thus found a secure starting-point for the full later development of the consecutive clause with the subjunctive, we may now turn our attention to the mood itself. Here the first and most prominent fact that meets us is, that the Latin language does not make any distinction between actuality and possibility, at least not in the clauses with ut, and not to any full extent in those with a relative. Where in Greek there is the sharp division into those with the indicative and those with the infinitive after wore, and in English a corresponding distinction between the fact after "so that" and the tendency after "so as to," in Latin we have the same subjunctive used for both alike. And it is safe to say that where there was no distinction in the means of expression, there was no distinction — at least no decisive, com- pelling distinction — in thought. This is all the more strange when we consider that the language is elsewhere quite careful to make this distinction between what is conceived of as fact, and what is conceived of as possible only, or likely, or sure to happen under certain conditions. More than that, even consecutive clauses themselves, when they are paratactic and not subordinate, also show this difference of mood with great clearness and consistency. If it was natural then to make this distinction in the language generally, the absence of it in the ut clause of result points to a The Subjunctive in Consecutive Clauses 87 certain lack of vitality in the mood of the clause, to a certain rigidity and ossification, which shows that the whole clause was no longer a free, living, sensitive expression of a shade of thought, but rather a symbol for a fixed concept. That the mood was the subjunctive, was due to the way in which the clause originated, but the precise mood had to some extent become a matter of indifference. Even in the clause with qui where both the indi- cative and the subjunctive are used, it is often almost impossible to see any difference between them. It is necessary, however, to guard against a misconception of these remarks. They do not in any way mean that the Romans were never aware of a difference between those consecutive clauses which expressed what had existence, and those which expressed merely a possible or conceivable state of things. Of course, they would see the difference at once if their attention was attracted to it. But the important fact in the case is just this, that in the absence of a difference in the mode of expression to bring out this distinction, their attention would not naturally be directed to it, and the lifelong use of such an undiscriminating form of speech, if we may use the term, would keep their attention, as a matter of habit, from turning to it spontaneously. We must perforce, there- fore, assume a national lack of sensitiveness to such distinction as was usually made by the moods, in this particular clause at least. The question how the mood could to such an extent lose its ordinary power of expression, is not by any means a difficult or hopeless one. If we examine the form of the consecutive clause in Latin, we are struck by the narrow bounds within which it is for the most part confined. The w/-clause depends in the great majority of cases upon a preceding ita, hoc, tantum, or a similar demonstrative word — a state of things which is found in an equally pronounced form in parataxis. The subordinate clause with ut seems in fact to have become, grammatically, nothing more than a modifier of this preceding demonstrative — so entirely subordi- nate to it that it is not really a separate statement of the result of the action in the main clause, but rather a complex adverb modi- fying, or rather explaining, a particular word in it. The whole clause, that is to say, is only a very elaborate ita or hoc. 88 J. J. SOHLIOHEE The explanation for this condition of things is not far to seek. We found that, in a sentence like ita non est validus ut haec faciat, the clause ita non est validus was, in the beginning, merely a more definite statement of the speaker's attitude toward the idea of the "man's doing this," than the subjunctive faciat alone was in itself; that is, the negative and the subjunctive both expressed the same attitude of the speaker toward the thought — the subjunc- tive vaguely and elusively, and the negative clearly and tangibly, and without possibility of mistake. Now, when such a vague and such a clear form of speech existed side by side in the same sen- tence, it can not be doubtful which of the two would be felt as the really significant expression. It was the negative in the main clause, the nemo est, the solus est, the non ita, and later the ita, hoc, tantum, etc., which received the emphasis, monopolized the speaker's and the hearer's attention, and thus secured the place of prominence in their minds. The subjunctive in the other clause was cast into the shade, beaten at its own tricks, as it were, and forced to the position of a somewhat inefficient and now rather worn-out piece of equipment, so far as the main purpose of the expression was concerned. In one sense this very obscurity and insignificance of the mood in the subordinate clause became its salvation. Like the insect in the amber, it preserved its form intact, and the con- secutive clause was one of the constructions which were least and last affected by the general process of decay into which the language fell during the last centuries of the empire. Though the mood in this clause may have been rather lifeless, however, it was still a regular form of the verb, which elsewhere in the language continued to perform its full functions of expression. Necessarily then it must, even in the consecutive clause, have continued to produce a rather definite impression on the hearer's mind, even if that impression alone no longer con- tributed in a decisive way to the comprehension of the speaker's thought. In the original form of the clause, with the negative antecedent, the attitude of the speaker toward the idea expressed by the subordinate part was a negative one, pure and simple. A sentence like this, for example, "The house is not so large that The Subjunctive in Consecutive Clauses 89 there is room for all," amounts really to an expression, on the speaker's part, that he is not able to accept the idea "that there is room for all," which is therefore, so far as he is concerned, a foreign element in his mind, as much so as would ever be the case in indirect discourse. And if we are correct in our view of the matter, this feeling of the foreignness of the idea expressed by the subjunctive must, in some degree at least, have survived even when the mood was used in clauses with a positive antece- dent. That is, even when we remove the negative from the sentence given above, so as to make it read, "The house is so large that there is room for all," the idea "there is room for all" is still felt by the speaker to be a thing apart from his own active thought, a sort of external standard, by which he is measuring his own idea of the house. In other words, the idea "there is room for all" is a more or less complete unit, which has a meaning and a value quite independently of his own thought and experience — a sort of abstract concept, which he may use as an element in his thinking and may comprehend, but toward which he stands in no personal relation whatever. It is the product of the thinking of other people, and is felt to have a more or less abiding force as a consequence of its existing in their minds. When Plautus says of the Menaechmi, ei sunt natei filii geminei duo, ita forma simili puerei, ut mater sua non internosse posset quae mammam dabat, neque adeo mater ipsa quae illos pepererat, he is stating his view of the degree of their likeness in terms of a general concept, "their own mother did not know them apart," which has a certain fixed value, much like a complex adverb, in the minds of the audience as well as his own, even before the words are uttered. If one says, "There are people who deceive their friends," he is merely asserting the existence of such people as are described by the terms of the general concept "who deceive their friends," which likewise the hearer has in his mind, fully grasped and clearly defined, beforehand. It is this feeling on the speaker's part, that in the consecutive clause after ut and qui he is not resting upon his own thought 90 J. J. SCHLICHER exclusively, that justifies the use of the subjunctive after positive as well as negative antecedents. For this is the mood in Latin which expresses ideas that are not felt to be originally one's own. The suggestion of foreignness is,.of course, not always equally strong, but varies widely, just as it does in indirect discourse. There is much the same relation, as we have already stated, between the ordinary positive form of consecutive clause and the stronger, original type with a negative antecedent, that exists between the ordinary subjunctive of indirect discourse, which merely assigns a quotation to its author, without explicitly accepting or rejecting it, on the one hand, and the absolute denial of responsibility for the foreign idea, and the positive refusal to accept it, on the other. We have so far pursued the main line of development in the history of this construction — the line which this development naturally followed when uninfluenced by other constructions. No form of speech, however, does or can altogether live such an isolated life, unless it has the exclusive use of the machinery of expression employed by it. So long as a mood, for instance, is associated with a number of different uses at the same time, these will, if conditions are favorable, necessarily exert an influence upon each other. So the consecutive clause with ut stood alongside a clause of purpose with ut. After certain verbs, like those of effecting, the clause might even in one time-sphere express a purpose, and in another time-sphere a result. We shall not be going too far in our claims if we assume that the complete monopoly possessed by the subjunctive in the expression of purpose must, under such conditions, have been a strong influence tending to establish and maintain a like monopoly for the subjunctive with ut in expres- sions of result also. That is, we can easily see how the consecu- tive clause with ut could come to have only a subjunctive, even though the consecutive clause with the relative had both the subjunctive and the indicative. The consecutive clause with the relative was for its part nat- urally subject to a number of influences which tended to ally it The Subjunctive in Consecutive Clauses 91 with certain other clauses introduced by a relative. For example, we observed the relationship and development of a series of these consecutive clauses, ranging from nemo est qui to sunt qui and multi sunt qui. It is evident that, while in nemo est qui the attitude of the speaker's mind toward the thought of the subor- dinate clause is absolutely negative, in cases like sunt qui it has become positive, and in multi sunt qui very much so. Now, as soon as we have a case where the reserve with which the mind gives its assent to a proposition is relatively small — that is, where the full acceptance of the proposition depends only on the removal of obstacles or the supplying of proper conditions — we have pre- cisely the sort of case in which the so-called potential is generally employed. And whenever a variety of the consecutive clause reached this point, it would inevitably fall under the influence of those forms of assertion that contained a potential and would, as it were, henceforth owe a double allegiance. This alliance of the con- secutive clause with a use of the subjunctive in which the mood still possessed a vigorous independent life and could still adequately express an idea in an independent statement, could not help but have a reviving effect upon the debility which was overtaking the mood in these consecutive clauses. The subjunctive might thus even here again be felt to have some independent sufficiency as a means of expression in itself, and would continue on a career of its own as a subordinate subjunctive of limited assertion. It will be observed, however, that it would not then be possible any longer to draw a sharp line between a consecutive clause which makes a limited assertion, on the one hand, and the subordinate potential clause, on the other. Indiana State Normal School Terre Haute, Ind.