132

is the simple, matter-of-fact tale that Plutarch tells, possessing, you may think, no special hint of priestly craft up to this point, at any rate. For it is surely open to us to explain the phenomenon precisely as we should the similar one today when a log on the hearth that is apparently extinct suddenly exhibits a vital flame. The context, however, may make a questioning reader pause and ponder; for immediately before this narrative we have the statement that Cicero's reputation with the populace for courage and decision was none of the best, while, immediately after, we find an exactly opposite characterization of his wife Terentia. Plutarch says that she was neither tender-hearted nor timid, but, on the contrary, a woman keen for distinction, who, according to her own husband's description of her, was much more eager to claim a share in his thoughts on politics than communicate her own to him on household matters. On this occasion, she went straight to him with the news, as she had been bidden, and spurred him against the conspirators. So we may ask ourselves the questions: Did Vesta really intervene at a crucial time in Rome's history, and quicken a dead fire miraculously, or, to use modern language, did Dame Nature chance to play a freakish trick at just the opportune moment, or shall we rather say that virginal hands with patriotic motives craftily resurrected the flame in a way that was known to them alone? This last explanation is, to my way of thinking, more than a remote possibility.

40

UNIVERSITY OF PENNSYLVANIA.

1.1.7.

REVIEWS

The Unity of the Latin Subjunctive: A Quest. By Edward A. Sonnenschein. London: John Murray (1910). 60 pp.

The theory of the Latin subjunctive which, for the last half-century, has enjoyed the widest acceptance may be stated as follows: The Latin subjunctive is formally a fusion of two Indo-European moods, the subjunctive and the optative. The IndoEuropean subjunctive had two broad general forces. First, it was volitive, the mood of will; secondly, it denoted futurity, pure and simple. The Indo-European optative was, first, the mood of wish; secondly, it was a potential, expressing the notions conveyed by our auxiliaries, 'should', 'would', 'may'. Latin, as a fusion of these two Indo-European moods, is held to have inherited at least three of the four notions above enumerated-the volitive, the wish, and the potential uses. As I aimed to show in Volume I of my Syntax of Early Latin, this theory admirably fits the facts of the Latin language, accounting for the use of the subjunctive not only in principal sentences but in subordinate clauses as well.

The book of Professor Sonnenschein, who fills the chair of Classics at Birmingham, challenges the view above stated and seeks to supplant it with a new one. To his mind, the Latin subjunctive possessed not several functions but one. It was the mood of "nonethical obligation". From this fundamental force he endeavors to explain all the uses familiar to us in Latin syntax. While he approaches his task with a competent knowledge of the extensive literature of his subject, while he argues his cause with a frank recognition of many of the difficulties likely to be urged against his views and defends his own position with admirable skill and acuteness, yet I am not persuaded that his interesting brochure will win many converts to the new doctrine.

The author takes up at the outset the objection that the Latin subjunctive might naturally be expected to have contained different basal meanings as a result of its composite origin-Indo-European subjunctive and optative. To this objection, he urges that there is no a priori reason why the IndoEuropean subjunctive and optative were not identical in function, precisely as in Greek the long vowel and the short vowel subjunctives and the first and the second aorist are equivalent in function. He admits that in Greek, though not invariably even here, subjunctive and optative are sharply differentiated. But he suggests that this distinction, which is shared also by the Sanskrit, may be purely the result of a primitive Greco-Indian unity. For IndoEuropean, Sonnenschein holds that subjunctive and optative alike expressed primarily this notion of nonethical obligation.

It is of course obvious that actual instances of Latin subjunctives denoting obligation are not nu

merous. We have two recognized types of the usage: interrogative expressions like hunc hominem non defendam, non diligam? cur abeam? and imperfects like haec mihi praediceres, 'you ought to have told me this beforehand'; haec tale ne faceres, 'you ought not to have done such a thing'. These, however, are infrequent and have long since been adequately explained as psychologically natural developments of the volitive use of the mood. It is doubtless from an appreciation of the paucity of subjunctives denoting obligation that the author endeavors to augment their number by interpreting as subjunctives of obligation certain examples ordinarily otherwise explained. A typical instance is Plautus, Trinummus 496, ubi mortuos sis, ita sis ut nomen cluet, which Sonnenschein translates, 'when you are dead, you are bound to be dead'. In view of the extreme unnaturalness of the English, it is difficult to subject to fair criticism a rendering like this. But I think it unlikely that the average student will recognize in the passage a notion of obligation, either ethical or non-ethical, either natural or logical. In my Syntax of Early Latin, Vol. I, 200 ff., I have considered at some length the example just cited, along with many others of the same sort, and venture to call attention to my own explanation. To my mind, sis in the Trinummus passage is merely a faded out 'would' potential. Another typical example of the same usage is Miles Gloriosus 689, hoc numquam audias, literally, 'you would never hear', but virtually 'you never hear'. So Most. 148, nec quisquam esse auxilio queat, literally, 'no one would be able', but in reality 'no one can'; And. 460, fidelem haud ferme mulieri invenias virum, ‘you would scarcely find', i.e. 'you scarcely ever find'; Cato, De Agricultura 17.1, id semen legere possis 'you can gather'. In the work cited I have given no fewer than forty examples of precisely this kind. Scotch English in particular abounds in similar instances of this use of the 'should', 'would' potential with the value of the indicative, e.g. 'How old would he be?' 'He'd be five and forty'. So also in Sanskrit and Greek we find the same use of the potential; see Delbrück, Vergleichende Syntax, 2. 371; Blase, Historische Grammatik der Lateinischen Sprache, 3.123. Compare also the German, ich wüsste nicht ich weiss nicht. Professor Sonnenschein has himself seemed to recognize this transition in another connection. On page 46, speaking of the 'should' 'would' potential, he says: "in proportion as this shadow cast by the condition becomes light, the expression of what would be approaches to an expression of what will be". He has himself, therefore, made easy for us the interpretation of the Trinummus passage which I suggest in my Early Latin Syntax, 'when you have died, you are dead indeed'. See page 200 for numerous convincing examples of the same kind. I must leave it to the judgment of the reader whether Sonnenschein's in

=

terpretation of these same passages (so far as he cites them) is a natural one. To me it seems invariably very forced. Thus he translates Terence, Adelphoe 254, abs quivis homine, quom est opus, beneficium accipere gaudeas, by 'one cannot but rejoice'. In this last example one seeks in vain for any notion of obligation either in the Latin or in the English. Sonnenschein's rendering, to my mind, expresses necessity (a different thing from obligation, natural, moral, or logical-Sonnenschein's three divisions). But the Latin, I feel, expresses neither necessity nor obligation-it gives merely a general statement of fact, 'one rejoices'.

The author's attempts to derive from the notion of obligation the various well-known and frequently exemplified uses of the subjunctive are most unsatisfactory. Thus he tells us that vendam 'I am to sell'; vendas = 'you are to sell'; vendat = 'he is to sell'. But I am to sell', etc., do not denote obligation. Nor is it easy to see how the notion of obligation could develop the notion which we call volitive, i.e. how 'you ought to go' can develop the meaning 'go!' Again Terence, Hautontimorumenos 273, mane! hoc quod coepi primum enarrem, conveys to my mind a very strong notion of resolve, 'I'm determined to finish telling'.

Similarly the development of the wish idea from that of obligation seems to me an incredible process. For the present subjunctive in this use Sonnenschein can cite nothing stronger than the German analogue er soll leben = vivat, the psychology of which is not clear. For the past tense the Greek peλov éX0eiv shows how a past expression of obligation might come to designate an unrealized desire.

Coming to the 'may' and 'can' potentials we find the gerundive cited in support of the possibility of a transition from the notion of obligation to that of possibility, e.g. homo non ferendus. But it is only in combination with a negative that the gerundive develops the 'can' notion. For affirmative sentences we have nothing either in logic or in usage to support the development.

Even less convincing are the explanations proposed for the origin of the subordinate uses of the subjunctive. Thus Plautus, Amphitruo 985, nec quisquam tam audax fuat homo qui obviam obsistat mihi, is translated, 'no one would be so bold that he shall stand', in illustration of the origin of the result clause. But I am at a loss to detect here any notion of obligation either in the Latin or the English 'shall', which seems to me to convey the notion of futurity merely and which moreover fails to bring out the real idea of the Latin-'so bold as to stand'. A similar arbitrary rendering of examples follows on page 37. So far as I can see, no one of these remotely contains any notion of obligation. I feel constrained to maintain this, despite the author's plea that the 'shall' of his translations is not a 'shall' of futurity.

Other types of subordinate clauses considered by Professor Sonnenschein are conditional sentences and the cum-constructions. If we follow him, we are to believe that in a sentence like si mihi obviam veniat, eum verberem, if he should come my way, I should beat him', the verberem "contains the idea of ought . . . marking the conclusion as the necessary consequences of the premises"! The way in which the force of the protasis in 'should' 'would' conditions is deduced is no less fanciful and unnatural.

In discussing the origin of the cum-constructions, the author fails entirely to explain how it is that quom took the indicative up to Cicero's time and then began to be construed with the subjunctive. By the terms of Sonnenschein's theory, we should naturally believe that cum Athenis essem started out with the meaning, 'when it was my duty to be at Athens', which then somehow came to mean, 'when I was in Athens'.

On the whole, Professor Sonnenschein's theory makes extreme demands on our credulity. If obligation was the basal meaning of the subjunctive and optative in Indo-European, why is it that clear instances of the usage do not abound on every hand? Why do we not find veniam, 'I ought to come'; credamus, 'we ought to believe', etc.? Sonnenschein in his explanations of development freely assumes these meanings (e.g. in the apodosis of conditional sentences, as laeter, 'I ought to rejoice'); but outside of imperfects (e.g. praediceret) and interrogative sentences (e.g. cur abeam?), where do we actually find these meanings in the Latin subjunctive? They simply do not exist.

Again, if the basal meaning of the Indo-European subjunctive and optative was that of obligation, why is it that we find no trace whatever of this force in the Greek and the Sanskrit subjunctive or in the Greek and Sanskrit optative? Can it be possible that the basal meaning of two moods disappeared absolutely in two languages whose oldest literatures are so well preserved for us? Nor can I find that a single other Indo-European language exhibits traces of the obligation idea, although Iranian has both subjunctive and optative, while Germanic, Slavic, and Celtic have either subjunctive or optative. Such a state of affairs, I repeat, is nothing less than incredible, if Sonnenschein's theory

has put it, the religious question shares with the food question and the sex question alone the distinction of having vitally interested all sorts and conditions of men from the beginning of historic time. In his position that the direction taken by the philosophical views of the Greeks was very largely influenced by the religious conceptions that were current before the rise of philosophy, Mr. Cornford is, accordingly, advancing neither a novel nor a startling view. Rather is he making a serious and, to a degree, a successful attempt to trace more precisely the lines along which Greek philosophy was influenced by religion. And he views religion from a distinctly social angle. He says himself that he is carrying into the domain of philosophy the same principles of interpretation that Miss Harrison has employed in the study of religion in her Themis, published somewhat earlier in the same year in which his own book appeared. The two scholars, in fact, wrote with each other's work in view. Mr. Cornford contributed a chapter to Miss Harrison's volume and Miss Harrison helped in the revision of Mr. Cornford's book. At the base of both works lies the thought of the newer French sociologists and psychologists, Lévy-Bruhl, Durkheim, and Bergson, with their emphasis on the social origin of religion.

The framework of Mr. Cornford's book is a distinction sharply drawn between two types of Greek religion and two corresponding and derived types of Greek philosophy. In religion, the Dionysiac type is set over against the Olympian; the one is collective (the thiasos is its essential), the other individual; the one rests upon a temporal, the other upon a spatial conception; the one practices sacrament, the other gift-sacrifice. For the one maintains that God and man are akin and that man can share in God's nature or even become one with him; the other upholds the superiority and aloofness of God and brands as insolence any least attempt to bridge the gulf that yawns between them.

In similar and corresponding contrast stand two schools of philosophy, the Italiote and the Ionian, the mystic and the scientific, the one clinging to unity at any cost, even that of refusing the evidence of the senses; the other tending toward a dead atomism, and eventually eliminating gods from the universe or relegating them to a quasi-existence of inactivity. It is the author's task to trace the influence of the two types of religion in molding the two schools of philosophical thought.

The first chapter aims to show that Moira, the spatial disposition of the elemental provinces, had once been of religious significance. The sons of Zeus had arranged among themselves a spatial division of the universe and thus departmental ordering of the world had appeared in religious representation long before it found its way into philosophy. It is even antecedent to religion; it goes back to a pre-religious stage and has a distinct

social significance. The relation of Anaximander's ǎrepov to the four subordinate elements corresponds to that of a totemic tribe containing four clans. The departments "are clearly marked off by boundaries of invisible taboo, each the seat of a potency which pervades that department, dispenses its power within it, and resists encroachment from without". The religious significance of spatial division of the universe points back to a stage when that division was continuous with the moral and social structure of the human group, and such divisions are moral, as in Anaximander, because those of the tribal group are moral. The tribe, for example, is divided into two exogamous phratries and this is the focus of intense religious and moral emotion. The form in which this appears in philosophy is the strife of opposites, and the myths make it clear that the sex contrast is the prototype of these contrarieties.

What are nature, the soul, God, viewed from the sociological standpoint? Nature is a living and divine substance (physis); it is at bottom the collective functions of a group, considered as a vital force proper to that group, the exercise of group functions manifested in magical operations. It is a vehicle of sympathetic interaction supposed to exist between things that are akin. It is conceived as the blood of the group kin. Soul is a representation of the collective life of the group. This collective life the individual recognizes to be something more than himself, i.e. it is superindividual and, so, superhuman. It is projected into a god. Gods are thus a development of souls-projections into non-human nature of the representation of the group soul. By this projection the non-human is made something more than a natural object. The fire is not wholly amenable to human control; it seems to have a nature and a will of its own. This will is a firedemon, the soul, not of an individual object but of a group of objects, in this case, of all fires. Four types of demons may be distinguished: (1) The genius of a social group, its representative, depository of its mana, developing into its KING; (2) or becoming confused with the local demon of fertility and developing into a HERO; (3) The collective soul of a magical fraternity, developing into a MYSTERY GOD; (4) The demon of a natural element, becoming an OLYMPIAN GOD.

The sharpest contrast is drawn between the mystery god and the Olympian, because the distinction between them is fundamental as well for the understanding of Greek religion as for the tracing of philosophy as it pursues its two diverse tendencies. The mystery god is characterized (a) by his thiasos, the group from which he has been projected; (b) by the temporal character of his functions, for he is a year demon, while the Olympian presides over a spatial department of nature; (c) by the magical, fertility-inducing ceremonies of his cult; (d) by his

retention of a degree of humanity that renders it possible for man to hold communion with him by the rite of a sacrament; and (e) by the monotheism, mysticism, idealism of his religion. The mystery god lives, because he is warm with mana; the Olympian isolates himself from man, first on Olympus, then in heaven's height; he cuts loose from his spatial province, as the growth of scientific knowledge reveals his uselessness therein, and his mana, or most of it, goes to local deities, which Olympianism failed to supersede but never recognized.

To these two theologies correspond two philosophies, and to the development of this correspondence the latter half of the book is devoted. To one who lays claim to no special skill in Greek philosophy, and has moreover already erred on the side of fulness, the temptation to touch only very lightly upon this portion of the book is too strong to be resisted.

The difference between the two philosophies depends upon their handling of physis, the "ultimate matter as yet undistinguished from soul', which is the datum of Greek philosophy. Ionic philosophy with its governing principle of Moira, the distribution of the world into spatial provinces, emphasized the spatial, geometrical aspect of this supersensible soul material, conceived as material substance filling space, emphasized it to the exclusion of soul and God. Its logical outcome was atomism. The atomists reduced physis to homogenous matter with none but spatial properties.

The mystical Italiote tradition, on the other hand, is inspired by a living religious faith that will retain God in its system at any cost. While the scientific tradition exhibits a more or less steady development in one direction, the various mystical philosophies are best viewed as a series of not necessarily consecutive or dependent attempts to translate a certain view of God into terms of a physical system. The concepts of time and number succeed that of space, and Dike, the periodic course of nature, replaces Moira, its spatial division. Derived from Dionysiac religion, with its annual resurrection of a year demon, it upholds the doctrine of palingenesis. The life of nature—man, animal, and fruitpasses through a yearly cycle; no change can be progress in a straight line, but it must follow the curve of the seasons.

It is a fresh and stimulating book Mr. Cornford has given us, but one which the reader instinctively adorns with marginal interrogation marks. This is not necessarily a defect, it may even be a virtue. One serious defect, however, the book certainly has. Mr. Cornford is evidently ill-read in contemporary work in the same field. For instance, he accepts unquestioningly Professor Burnet's view of the metaphysical character of physis, and bases much of his argument upon it, considering physis in this sense the datum of Greek philosophy, apparently unaware

AN APPEAL AND AN INVITATION

Repeatedly during the last ten years there has been an urgent call to unite the classical forces of Philadelphia and its vicinity into some sort of Association that should strengthen their cause in every way possible. At the request of several prominent teachers, the President of the Philadelphia Classical Club, the largest existing combination of local classicists but composed only of men, appointed last November an organizing committee from that body which, after enlargement through the addition of others who were not members of the Club, should undertake the task. This seemed to be the most impersonal way to initiate the project. Care was taken that the men and the women selected should represent justly and adequately public and private schools, colleges, and normal schools both within and outside of Philadelphia. On the committee are representatives both of teachers of the Classics and of those who while teaching a different subject or not teaching at all are staunch supporters of Greek and Latin. This accords with the basic spirit of the entire undertaking, which is not for the aggrandizement of individuals or institutions but to promote in our system of education liberal studies and more particularly the Classics. Those that eventually serve the Association as officers can look forward only to self-sacrificing labor and the satisfaction of fighting shoulder to shoulder for a cause for which they could formerly contend only impotently as individuals.

Greek and Latin are essential elements of a cultural education for a large number of boys and girls in every community. At present their legitimate claims are not satisfied nor even fairly considered in many quarters, partly because public opinion is not sufficiently instructed, partly because the rightful demands of the majority have been pressed so dominantly and noisily that the voice of the minority is barely heard. To effect a proper change in public sentiment, we must gain the active alliance of all citizens who are grateful for the classical training that they received in their youth and desire it to be duplicated in the present and in coming generations that a certain type of cultured gentleman and intellectual leader may not perish from Ameri

In a paper entitled Hepi Puoews, in the Proceedings of the American Academy of Arts and Sciences, Volume 45, Number 4.

For this paper, entitled On Anaximander, see Classical Philology 7.212-234.

can civilization. This appeal, therefore, is to a much wider circle than those who are merely professionally interested.

The activities of the Society will naturally assume various forms. Thus, the attention of the general public can be gained by such exhibitions of 'Living Latin' as are now in vogue in many enterprizing schools, and by the presentation of the life, thought and material remains of antiquity in attractive lectures, illustrated by the lantern, where advisable. Moreover, such discussions of scholastic topics could be arranged as would interest and invite to participation even the layman.

Much more obvious, of course, are the possibilities for the teachers who ally themselves with this movement. Besides the mental stimulus that comes from every new contact with the scholarship and the pedagogical skill of others, there is the social pleasure and advantage of knowing better every year other workers in one's own field. Then, too, the best source of information about professional vacancies and the candidates available for them is among one's friends. Young teachers in particular would profit by a wider acquaintanceship. As a rule, the individual who subscribes to no classical journal, and never lends a hand at any meeting of fellowworkers, but with self-satisfied selfishness performs merely his routine of daily work, pays the penalty by obscurity, intellectual stagnation and pedagogic inefficiency. In view of the fact that it is commonly the overworked men and women who are most active in all that makes for the common good, the plea of being too busy is rarely an acceptable excuse. Nor can even that other useful excuse of poverty avail in this instance; for the membership fee of the new society will be hardly more than nominal. In the neighborhood of Philadelphia, therefore, the isolation of any classical teacher or of any friend of Greek and Latin will in the future be wholly wilful. There are plans for meetings of a largely social character, such as luncheons, informal receptions, round table conferences and an annual dinner, and a determined, organized effort will be made to make all who attend any of these cordially welcome and rapidly acquainted with one another.

Under the able guidance of the Chairman, Professor Walter Dennison, of Swarthmore College, the Organizing Committee and its sub-committees have held many lengthy sessions, and can now definitely announce that the first meetings of the new Society will be held on the morning and afternoon of Saturday, March 14, at the Drexel Institute in Philadelphia. The exact program will be published in a later number of THE CLASSICAL WEEKLY. Any suggestions as to its character or with reference to the lines of endeavor that the new Society shall eventually pursue will be gratefully received by the above mentioned Chairman.