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. BOOK REVIEWS Kritisch-historische Syntax des grtechischen Verbums der klass- ischen Zeit. Von J. M. Stahl, Professor an der Universi- tat Mtlnster i. W. Heidelberg, 1907. Pp. xii + 838. M. 20 (gebunden 21). This monumental book belongs to the " Indogermanische Bibliothek" grammars edited by Hirt and Streitberg. They are uniform in style. But the margins which sufficed for a thin book of two hundred pages, like Streitberg's Gotisches Elementarbuch, helped out by heavy paper and type mainly large, are insufficient for a ponderous volume of over eight hundred pages and are in fact even reduced, under the difficulties of trimming so thick a mass. The paper is lighter, and the proportion of fine type, much of it heavily packed with Greek examples, and some- times running consecutively for several pages, is very large. The result is forbidding to the eye. The book should have been put into two stately volumes, with larger type and wider margins. It well deserved such generous treatment. It is a work of enormous industry, and presents the reader with a great storehouse of examples, for which he cannot be too grateful. In any case, the volume should have been provided with running headings at the tops of the pages, corresponding to the divisions given to the table of contents on p. xii. As it is, only the main divisions are given, so that, for example, the heading Lehre vom Modus covers 376 pages with nothing whatever to guide the reader, outside of the packed text itself. It is much to be hoped that in subsequent editions this defect may be remedied. The smaller books of the same series are rightly equipped in this respect. Only a very long review could deal with the details of this book. The present one must therefore be concentrated upon the governing system. In the preface, the author states that a satisfactory historical treat- ment of Greek syntax did not yet exist when he began his work. This is true. He also states that the critical standpoint has not been taken, meaning that a fresh judgment must be passed at many points upon the readings of our texts. I find that in his actual procedure he gives his syntactical theories too much weight, for example, in the case of tva . . . . aX(f>oi, Od. xvii. 250, which I not only accept but welcome, while Stahl emends, against all the MSS. 439 440 Book Keviews Likewise in the Preface, Stahl says that he takes from Delbriick his view of the original forces of the Subjunctive and Optative (respectively Will and Wish), and from Windisch the view that the relative was originally an anaphoric demonstrative; but that he has made these views his own by putting them on firmer foundations. He has not felt it necessary to go completely through the mono- graphs that deal with Greek syntax, but believes that nothing important has escaped him. He explicitly acknowledges assistance from Schanz's Beitrage only. Except as thus shown, and in a few cases where the author's text- criticism is defended by the agreement of others, I find no references to other work on syntax, even of recent date. There is great convenience in this plan, and it may in time become the only possible one ; but it does not wholly satisfy at present, especially in a book that allows itself such a compass. The opening chapter is devoted to a discussion of method, of which only a part can be mentioned. Stahl enumerates the empirical method, which observes and classifies phenomena (as, e.g., that idv takes only the Subjunctive), the logical method, which deals with the especial contents of thought, the historical, the comparative, and the psychological. The empirical and historical methods are able only to acquaint us with the external forms of syntactical usage. The logical method penetrates to the inner laws, without being competent, however, to exhaust the nature of speech-usage. This lies in the character of speech itself, which fol- lows logical laws no more exclusively than life does. Only the psycho- logical method will enable us to unravel the processes of thought. This is in the main sound, but goes astray in leaving to the psycho- logical method only those phenomena in which the mind has worked wwlogically. Whatever the mind does, whether unlogical or logical, is psychological. The chapter deals also with the passing-over of a meaning into a new meaning, with specialization of meaning, with logical and formal assimilation, and the like. Next follows a chapter on fundamental conceptions, dealing with such matters as the definition of syntax, the sentence, subject and predi- cate, word-questions and sentence-questions, true questions and rhetorical questions, and the division of all sentences into those of desire and those of judgment. This division plays a constant part throughout the book. Most important, too, for the reader is the distinction between synthetic sentences, or those which are closely bound to the sentence on which they depend, and parathetic, which are not so bound, and are in reality independent. We must pass the chapter on "Voice," with its old-fashioned name ("Lehre vom Genus"). Book Reviews 441 la the chapter on "Tense," a number of delicate distinctions are made or attempted. For example, the Imperfect, which is defined as the past of lasting action (dauernde Vergangenheit), may be absolute (occupying length of time in itself), or relative (lasting alongside of another). The Aorist is defined as expressing the past by and for itself (Vergangenheit an und fur sich); but like the Imperfect, the Aorist, it is said, may be either absolute or relative. Since the Imperfect and Pluperfect express lasting action or state in the past, the expression of momentary single actions is left to the Aorist. But the Aorist may also be used of lasting actions, if the lastingness is unimportant, and there- fore does not need to be expressed (p. 124). In consequence, there are Aorists expressing actions which, as a matter of fact, covered a good deal of time. Thus Stahl endeavors both to correct and to follow a common erroneous opinion. To my mind, though it is a general belief, few ideas are so perverted in syntax as the one that the Imperfect (and Present in most uses) express duration, continuance, etc. It is a matter of complete indifference in the choice of tense whether a trifle of time or a great deal of time is occupied by a given act. I may say, of an instantaneous photograph of the flight of a shell, "the camera was snapped just as the shell was passing" (Imperfect); and I may say "the shell flew fifteen miles " (Aorist). The time covered by the second act is much longer than that covered by the first, but the first act is repre- sented in the going on, while the second is represented " in summary," " as a whole," to quote my definition for the Latin aoristic Perfect in the Hale-Buck Grammar 466a. In place of "lasting" or "continuous," the word "progressive" should be used for the Imperfect, as in most grammars of English. For the Aorist, no better name than this has been devised; for all others (even the name "eomplexive") include something the mention of which suggests limitations that should be omitted. The long chapter upon the moods opens with a definition: "The mood indicates the way in which the speaker expresses the verbal conception in its relation to reality" (p. 220). From this I dissent. The mood expresses the mood-idea, whatever it is, for itself, and makes no comparison of the conception with that of reality. The Optative of Wish expresses a wish, not a wish in its relation to reality. Stahl's phrase is a relic of the time when Subjunctive and Optative were thought of as deflections from the Indicative. The oldest force of the Subjunctive, Stahl finds, is that of Will, and of the Optative that of Wish. I think the conclusion probable, but the arguments not all sure. Thus the fact that the force of Will always remains in the independent sentence, while the Future force goes out, does not prove that the force of Will is the older. In Greek, the sub- junctive has driven out the Optative, in Germanic the opposite; but this 442 Book Reviews does not prove the contradictory proposition that each is the older mood. The future force of the Subjunctive arose out of the Voluntative. (I like my word Volitive better). The Optative, which originally expressed Wish, came also to express a Conception (Vorstellung) of the speaker (pp. 11 and 256), and then widened its meaning to express Conception in general, and from this came to express the Conception of another than the speaker; whence the construction in Oratio Obliqua. The processes by which the author thinks that these new forces came to be associated with the Subjunctive and Optative moods are described as follows: "The passage to the future sense in the Subjunctive was yielded by the fact that in Will a striving after future realization (Streben nach zukilnftiger Verwirklichung) is included." For the still more dif- ficult problem of the historical relation of the far-ranging forces of the Optative, we have only this (p. 236): " As now the Wish, not being accompanied by a striving after realization, belongs to mere Conception (der blossen Vorstellung), so the Optative in judgments also marks the Conception of the existence of something {die Vorstellung, doss etwas set), which at the same time can be expressed as the view or opin- ion of the speaker. The force of Wish is capable of weakening itself to that of mere Admission (' concessive Optative '). In this sense the Opta- tive is employed not only of willingness that something shall take place, but also of the acceptance of something as actual (Annahme, dass etwas sei). This is most indefinite, and, except for the derivation of the " con- cessive " Optative, shows no touching of the feet upon solid ground. Analyzed, it amounts to this: "Because a tw'sft-conception is expressed by the Optative, therefore a jwdgweni-conception will also be." It is a complete non-sequitur that is thus affirmed. The truth is that this whole vague theory of the Optative as the mood of Vorstellung, to be found not only here, but in most works on the Greek moods, or on the mixed moods or surviving moods of other languages, is, like the phrase, Streben nach Verwirklichung for the Subjunctive, an inheritance from the metaphysical school of syntax. I have told the story briefly in "A Cen- tury of Metaphysical Syntax," Publications of Congress of Arts and Sciences, Universal Exposition, St. Louis, 1904, Vol. Ill, and again, still more briefly, in " The Heritage of Unreason in Syntactical Method," Proceedings of the (English) Classical Association for 1907 (Vol. V), London, John Murray. 1 To make my criticism intelligible on its most 1 1 told it again at the Versammlung deutscher Philologen und Schulm&nner at Basel in 1907 ; but the highly condensed form in which all the papers of these meetings now appear crowded out everything, except a mere hint. Koppin's programmes, so much quoted, do not unravel the threads, and confuse historical order by failure always to use first editions as well as later ones. Book Reviews 443 important side, I must sketch the matter in part again, beginning with Gottfried Hermann. In his book De emendanda ratione Graecae grammaticae, 1801, Hermann, taking his hint from Hasse, fitted the Greek moods to Kant's Modal Categories (in the Critique of Pure Rea- son) of Existence, Possibility and Necessity, making the Indicative express Existence, the Subjunctive Objective Possibility (possibility in the nature of things), the Optative Subjective Possibility (possibility as thought), the Imperative Subjective Necessity, and the Verbal with -t«os Objective Necessity. What followed was a twisting and turning of these ideas and phrases. Matthiae, in two Greek grammars, 1807, 1808, threw the emphasis, for the Optative, on the second half of Hermann's phrase ("as thought"), and made it accordingly the mood of thought. The Subjunctive he also made to be a mood of thought, the difference between the two being that the Subjunctive expressed the act more definitely, as depending on external circumstances, the Optative less definitely. Dis- sen, "Habilitationsschrift," De temporibus et modis verbi Graeci, 1808, made the Optative express a conscious thought. The Subjunctive, which expressed an act as depending on the nature of things (external circum- stances), he made the mood of Conditionality. (Later, by a curious con- fusion, the Subjunctive was made to express, not something Conditioned, but Condition itself. Prom this comes down a great mass of error, by which all Subjunctive clauses, including those with «os, vpiv, etc., are made conditional; so in a long succession of German writers, and, in our English-speaking world, Goodwin, Allen-Hadley, Goodell, Babbitt, Monro Homeric Grammar, and even John Thompson Greek Grammar.) In 1812, Thiersch published two Greek grammars. For the Optative he accepted Matthiae's view, making it represent an act as merely thought (als bios gedacht) as an idea, a conception (Vorstellung). Thus a clause expressing a repeated action in the past must be in the Optative, because the acts did not really happen together, and the putting of them together is an act of the mind. For the Subjunctive Thiersch accepted Dissen's view, combining all the phrases used about it in Dissen's paper. The Subjunctive is the mood of the dependent, the conditioned, the uncertain. Thus in ioy*£y, " let us go," the going is conditioned by the will of the person addressed, and the mood must accordingly be the Subjunctive. Thiersch recognized Will as the force of the Subjunctive in what he calls its earliest seat, and derives the future force from this, on the ground that acts lying in the future depend either upon the will of some one, or upon the later course of events (dem weitern Erfolg). In this connection, he uses the phrase, wo dieser noch zu erwarten, " where this is still to be expected." Out of this has grown the use of the technical name Erwar- tung, now generally employed in Germany for one of the two forces of the Greek Subjunctive. Different books that followed made different combinations. In the 444 Book Reviews application to other languages (for this at once took place), Matthiae's system triumphed. The Latin Subjunctive, as corresponding to Mat- thiae's two moods of thought for Greek, was made the Mood of Thought (so Schulz, 1825; Zumpt, 1818, Madvig, 1844, etc.). Jacob Grimm adopted the same scheme for German, Maetzner for English, etc. It became, and in the main still remains, the dominant system. The Op- tative of Greek, the Latin Subjunctive, the German Optative, etc., express blosse Vorstellung, subjectiven Gedanken (Gerth-Kiihner), une simple conception de Vesprit (Antoine), eine gedachte, vorgestellte Handlung, einen Gedanken (Methner). To the same source, too, Schmalz Lat- Gram? owes his "Active" power of the Subjunctive (taken from Latt- mann), from which he develops the power of Subjectivity. Even Del- brtick Conj. u. Opt., 1871, employs Gedankenwelt, Vorstellung, etc. (though he derives this use from the Potential, not vice versa) to explain several dependent Optative constructions in Greek, including Oratio Ob- liqua; while he makes it the general basis of his entire treatment in Der germanische Optativ im Satzgefilge. It is all (except some of Thiersch's sound observations about Will and Expectation for the Subjunctive) the baseless fabric of a dream. Its source alone, the direct application of a passing metaphysical system, should condemn it. But, in any case, it is inconceivable that language should have begun with such abstractions. A given mood must have expressed, in the beginning, a fairly definite meaning; and it must have got its power of expressing other meanings through a series of natural associations. It is the business of a syntacticist, first, to register the meanings found, and then to endeavor to trace the actual historical sequence of these meanings. The other phrase mentioned, which plays a large part in many modern systems, goes back to Baumlein, Untersuchungen, 1846. Baumlein makes the Optative the mood or the purely Subjective, which is really the same thing as Matthiae's twist of Hermann's phrase. The Subjunctive Baumlein makes express (p. 177) Tendenz zur Wirkliehkeit. On p. 35, where the definition first occurs, he says that it expresses das Streben, das Tendenz zur Wirkliehkeit. Though this definition is not founded upon any statement of the metaphysical school, it is only a mind trained in the thinking of that school that could have originated it. It would seem as if Baumlein were looking for some phrase which should cover the two forces in the subjunctive distinguished by Thiersch. Such phrases, framed to cover differing meanings do harm rather than good; for the conception of the mood, thus crystallized, is then used to explain the different uses of it, as it is, for example, by Stahl. His sentence on p. 242, end of 2, is simply a condensed statement of the Hermann-Matthiae-Baumlein doctrine: "The three moods accordingly express a subjective view of the speaker, within which something may Book Keviews 445 appear as agreeing with reality, striving toward it, or existing outside of it in the conception." Again, his phrase Subjective Mdglichkeit oder Denkbarkeit of p. 264 is an absolute reproduction of leading phrases of Hermann and Matthiae. And his forcing of all Potential Optatives into the " Subjective " meaning (e. g. " what mortal man could recount them all," Od. iii. 114), as against the "objective" meaning of Svvarax, is of a piece with the forcing practiced by the same men. To my mind, our first business in mood-syntax is to get rid of all this metaphysical inheritance, which belongs to an unscientific age, and to study language directly, basing our conclusions upon observation of forces, and not upon tradition. The first exhibition of examples following the exposition quoted for the Optative is divided into three groups, the Optative of Wish, the Concessive Optative, and the Optative of Conception. Here, if anywhere, the clearly distinguishable notions that can be conveyed by the Optative ought to have been given,— the leading forces of the mood. This is not done. Later, however, some of these appear as, e.g., the Potential Optative (p. 264. 2) and the Affirmative Optative (p. 264. 1) which I am glad to see, while still preferring my own phrase Optative of Ideal Certainty which excludes the Potential Affirmation, as Stahl's term does not. Rarely is there any translation. To my mind, one or two examples at least, under each category, should be translated, even in a book like this; to enable the reader to know precisely what mood-idea the writer himself attached to that category. The bare title " Conception " tells us nothing except " not-wish." The exposition of the forces of the two moods is intermingled with the treatment of the functions of ov and «e. Our views about these particles, says Stahl (p. 253) must be based upon the usage in independ- ent sentences. From these it appears that the particles are never used in true sentences of desire, and may be used, or not used, in sentences of judgment. It follows that, when used, they add nothing to the meaning, but only stamp the expression as a judgment. "This must necessarily come about (p. 255) through the fact that, as against the merely postu- lated, the act expressed possesses reality somehow or other (irgendwie Realit&t besitze). The modal particle raises the expression to assertion, its nature is subjective affirmation" (the emphasis is Stahl's). Here is the keynote of his system for the particles. It is not only mechanical in itself, but it breaks the rest of the system down. It plays havoc with the definition given of the Optative, and the distinction between this and the Indicative. The Indicative expresses reality, and the Optative, which we have been told (see quotation above from p. 242. 2) expresses something as existing in the Conception outside of reality nevertheless " somehow or other possesses reality" when it is accompanied by &v or k«! The same principle is applied to synthetic (i.e., really dependent) 446 Book Keviews clauses. Thus in Subjunctive Clauses the assertion of a future reali- zation indicated by the use of the modal particle becomes Supposition (wird .... zur Voraussetzung, p. 258. 1). The passage from Assertion of something as future to the Supposition of it is a very easy one ( Uegt sehr nahe, 258. 2); for whoever asserts that something is going to be or to take place, supposes (assumes) its existence and occurrence. To my mind, this is a mere exercise of apparent ingenuity to maintain a position falsely taken elsewhere. The function of assertion is an entirely differ- ent thing from the function of assumption. The very fact of assuming something shuts out the asserting of it. The same explanation is given by implication (p, 259. 3) of the Sub- junctive with els o and ?«?. These, being accompanied by the particle, are judgments (that is, to put the matter in simple language, assertions) about the future. I find "assertion" to be entirely at variance with the feeling of these clauses. At times the explanation is right. Thus the final clause with pa and the particle av or kc is in (in origin) a statement about the future, as I pointed out in 1894, against Delbriick's treatment, in my Anticipatory Subjunctive in Greek and Latin. So, e.g., II. i. 32, could still illustrate what the paratactic stage was. Thus "do not anger me: by thus acting (is), you will go more safely." The same is true of the final clause with the relative. 1 But, even here, Stahl, in order to keep up his general formula, puts the matter too mechanically in saying (p. 260. 1), "the sentence of desire appears in the form of a judgment." A thing cannot appear " in the form of " something else which is totally different. I should say rather, "these two ways (action willed and action counted upon) in these combinations express ideas which are so close together that both may fairly be labeled Purpose. On p. 263, a somewhat better account of the modal particle is given, though the basis on which it is erected is false. " It appears from the foregoing (the use of av or « in conditions, with fo>s, etc.) that in Homer the modal particle rarely fails to be used, and that accordingly a strong need already asserts itself to stamp the sentence of judgment exactly as such. This even led to the result that people added the particle to the clause of purpose and the general clause, which belong to the sen- tence of desire, and so gave them the form of the sentence of judgment." If a strong need was felt to stamp the sentence of judgment (a state- ment) as such, then a strong need would naturally be felt not to stamp the general condition as a statement, because it is not one. It is true that the clause takes the form which, if it were independent, would indi- cate a judgment. But it is misleading to say this. To my belief, no iDelbrflek accepted my criticism, saying (Vergl. Synt. II. 368): "I accordingly confine myself to Greek, remarking in this connection that I now with Hale (Antici- patory Subjunctive) distinguish the Volitive and the Prospective Subjunctive." Book Keviews 447 thought of a judgment underlay the Greek consciousness in expressing the "whenever" idea, etc. The particle, as I have for many years taught, is purely formal, and due to a general leveling which resulted from a variety of causes — among which the one given by Stahl does not exist. The important question of the explanation of the use of /hj in condi- tions is intended to be managed in the statement (p. 767), "/xjj belongs naturally to sentences of desire and to those sentences of judgment which do not indicate a real act Tatsachliches, but only something merely supposed or assumed, namely final sentences and hypothetical sentences." This again is an attempt to make a hard-and-fast formula work. When we were dealing with dv and xe in the condition, we were told that the presence of the particle indicated the possession, somehow or other, of reality. There is reality enough for av, but not enough for oil In order to explain prj in the condition, the difference between state- ment and supposition is insisted upon. To justify ov, the necessary inclusion of the idea of statement in that of supposition is insisted uponl Further, Stahl's reasoning is founded upon the essential nature of thought, and so would everywhere be operative. What, then, would be left for him to do with the negative of the Latin condition, which corre- sponds in function to.oi, not to /uij? This illustrates a general fault which I have to find with his procedure. It violates the very principle which he lays down in the chapter on method : "The essential character of language is not to be got at by logical subtleties and abstractions." Again, Stahl violates his main purpose, the historical, where (p. 480 and elsewhere) he explains the Modusversehiebung as due to Oratio Ob- liqua, — an almost universal error, found, e. g., in Goodwin, Allen-Hadley, John Thompson, etc. The Optative of the Modusversehiebung is in its full maturity in Homer, the Optative of Oratio Obliqua is not. The sec- ond is therefore presumably the younger construction, and cannot be the mother of the older construction. The historical relation is probably the opposite one. It will be seen that, in spite of its great importance through the enormous amount of material collected and worked over, Stahl's book belongs, in my judgment, to the syntax of the past, not to the syntax of the future. My criticism would be more valuable if it were constructive also; but there is no space for this. I can only refer to my treatment of " The Origin of Subjunctive and Optative Conditions in Greek and Latin," in the Harvard Studies in Classical Philology, Vol. XII, to my views upon the development of the various powers of the Optative, in the Proceedings of the American Philological Association XXXII, pp. exx-exxii, 1901, and the Hale-Buck Latin Grammar (p. 239), and to my "Anticipatory Subjunctive in Greek and Latin," Studies in Classical Philology Vol. I, University of Chicago Press. W. G. Hale