Art. II.-GIOTTO AND EARLY ITALIAN ART.

A History of Painting in Italy. By J. A. Crowe and G. B. Cavalcaselle. New edition in six volumes. Edited by Langton Douglas, assisted by S. Arthur Strong. Vols i and ii. London: Murray, 1903.

IT would be hard to devise better words of welcome for this great work as now reissued by Mr Murray than those with which Mr Douglas speeds it in an opening sentence of his preface.

'Notwithstanding,' he says, 'all that has been done in the last forty years by archivists on the one hand, and by connoisseurs on the other, with the object of elucidating the history of the central Italian schools, this book still remains the standard authority upon the subject. Of genuine additions to knowledge,' Mr Douglas proceeds, 'of scientifically verifiable facts, accepted as such by all serious and intelligent students, how little has been added to that particular fabric of human learning which owed so much to Crowe and Cavalcaselle! Much that passed for knowledge a decade ago has been proved to be unfounded theory; and, were it not unwise to prophesy, we would venture to predict that, in the coming decade, the field of art criticism will be strewn with the wreckage of many other pretentious but cheaply built structures.'

It is probable, indeed, that there is no domain in which greater difficulty attaches to the differentiation of theory from fact, in which the subjective and objective are harder to distinguish, even for those most desirous of distinguishing them, than that which presents itself to the critic of early art. But art criticism is valueless unless its methods are scientific; and the very difficulty of achieving such a result renders the attempt more obligatory. Mr Douglas does well to emphasise so important a truth; and if he seems a little eager to anticipate the havoc which better methods may produce, it will appear that he is specially entitled to make the prediction.

Naturally, in the review of a book with the bulk of which the public has been long familiar, it is the element of novelty which claims closest attention; and such novelty, without a doubt, appears most obviously in the

share of the editor. It will not be unreasonable, therefore, to give his work the first consideration.

It is easy to perceive that the work of editing a manuscript which has not received the final revision of its authors involves exceptional difficulties. Any reader who does not shrink from the labour of collating the second edition of Crowe and Cavalcaselle's great work with the first may soon convince himself of the severity of the problems with which its editor has had to contend. Thus at the bottom of the sixth page occurs the peculiar statement that the face of Christ,' in one of the rude paintings of the Catacomb of St Callixtus, expresses some of the feeling which so nobly characterises effigies of this kind in the fourteenth century.' Can Sir Joseph Crowe have affirmed here the very parallel which, in the first edition, he went out of his way to deny? Such a conclusion will hardly be justified by an attentive perusal of the passage in which the statement occurs. The general verdict is so clearly the same as that given in the first edition that the editor would perhaps have been within his rights in correcting 'some' to 'none.'

An instance of still greater importance occurs on page 133 of the same volume. Sir Joseph Crowe suggests that Giovanni Pisano, 'before he went south, probably carved the celebrated group above the frieze of the eastern gate of the Campo Santo.' The editor points out that, like Morrona and Rosini before him, he has confused two entirely different works; that the Madonna and Saints of the Campo Santo is of a quite inferior order; and that the group by Giovanni is in reality above the frieze of the eastern portal of the Baptistery.' Here again a reference to the first edition (page 143) seems to show that the author's error amounted to nothing more than a slip of the pen. The old frieze of Bonamicus' he says 'on the eastern gate of the Baptistery was crowned by a standing figure of the Virgin and Child between two Saints,' and he proceeds to call attention in a note, first, to the inscription, which shows the work to be Giovanni's (quoted also in the second edition), and next, to the error of certain

*Not a trace,' in the first edition.

commentators (presumably Morrona and Rosini) who confused it with a different work on the Duomo.

Revelations of this kind are inevitably disconcerting, and create an unpleasant sense of insecurity with regard to the text as a whole. There can be little doubt that Sir Joseph Crowe intended the second edition of the History to supersede the first; but it is fair to question whether, in the first two volumes as at present published, that result has been attained.

Yet this is by no means the only question which the new issue will provoke. Sir Joseph Crowe, the editor informs us, was engaged until the year of his death in preparing the new edition of the History. It was only in 1896 that the author's death occurred. This is but a short time ago; and the interval seems of small account when compared with the forty years which have elapsed since the appearance of the first edition. Nevertheless, even in this short time, critic and archivist have been busy, and here and there, by dint of strenuous gleaning, have added a grain to the store of true knowledge. In this connexion again a task of the utmost delicacy presented itself to the editor. Except where documentary evidence of a decisive kind is brought to light, it must always be a nice question what degree of concurrence among living authorities is required for the transformation of a novel theory into accepted fact. Mr Douglas, in the passage already quoted from his preface, shows a complete recognition of this initial difficulty. The high reputation which Crowe and Cavalcaselle's History, in its original form, acquired, the added weight which must attach to their opinions as now reissued after long and mature reflection, give their work the strongest claim to consideration and respect. It would obviously be wanting in taste to allow theories which were mere theories to be appended to their text, or, in a work which will necessarily be of unique value to serious students of every nationality, to admit dissenting opinions which were unsusceptible of proof. The editor would naturally desire that his work should contain a summary of the latest results of research; he would of course feel it to be a misfortune if imperfect theories were disseminated under cover of the authors' reputation.

Opinions will no doubt differ as to the degree in which

the editor succeeds in keeping the suggested ideal in view. A typical instance of his method occurs in connexion with the important problem of the chronology of Giotto's early works. The subject is so interesting that it deserves to be treated in some detail; and we can hardly approach it better than by briefly reminding the reader of Crowe and Cavalcaselle's own account. The greatest care has been expended by Sir Joseph upon his revision of Giotto's life, but he has not found occasion to modify the main outline as he originally determined it. One change only is introduced, the date of Giotto's birth being now given as 1267 instead of 1276; his works follow the same order as before. The first are those of the Upper Church of Assisi, executed while Giotto was still young; later, in his manhood, he was called again to Assisi by Fra Giovanni di Muro, to paint the ceilings of the Lower Church. Without asserting it as a fact, the authors leave us no option but to believe that this call took place in 1296, as the artist, after painting the Allegories, and possibly also the scenes from the life of Christ, proceeded to Rome before the end of that year. He stayed at Rome six years the inside of six years, clearly-executing the navicella, the ciborium, and other works that have perished; and he went to Florence not later than April 1302, where he painted the chapel of the Podestà. His next known work is the Arena Chapel, executed in 1306 at the age of thirty-nine. Considering the monumental character of the achievement, the authors feel that they put no impossible strain upon Benvenuto da Imola's testimony that Giotto, when he painted it, was 'still fairly young.' This chronology is acknowledged by the authors to be tentative, especially so far as the Assisi works are concerned; and it cannot be denied that they take refuge in a certain vagueness of statement.

Yet the complexity of the subject may be easily deduced from the fact that the editor has suggested a new chronology in his notes, involving so revolutionary a conception of the artist as to be totally subversive of all the authors' observations. The editor reserves judgment as to the date of Giotto's birth; but with regard to the order of the early works he is dogmatic and precise. The first were at Rome; they were followed by the frescoes of the Bargello and by the St Francis series

in the Upper Church of Assisi (1302-1306). Next came the Padua period, and after it the Allegories and the Life of Christ of the Lower Church. The inversion is so startling as to be all but incomprehensible, till connected with the new theory proposed by the editor as to Giotto's early training, touching which he again feels himself compelled in his commentary to disregard the statements in the text.

The authors hold that Giotto was, by training as by birth, a Florentine. To them Cimabue-the great Florentine Homer, of whom the modern critic only knows, and is forgetting to respect, the name, but whose reputation was considered by Dante comparable to that of Giotto himself—is an artist with a style they can recognise and trace, foreshadowing already the superlative qualities of the nascent Tuscan school; they can even detect certain peculiarities of style in Giotto's early work, which they regard as the immediate result of his master's influence. The editor believes, on the contrary, that Giotto's training took place in Rome, and that Rome, not Florence, was the centre of the revival of painting in the thirteenth century. He regards Cimabue either as a myth, or else, 'like Giotto, artistically a scion of Rome.' Giotto, as is well known, visited Rome, but he did not (we are told) find assistants there, as has been hitherto supposed; he found a master. There is no proof' (says the editor, vol. ii, p. 99) that Cavallini ever assisted Giotto at Rome; the probability is that the younger master assisted and was influenced by the older.'

This theory provides us with a key to the new chronology. In the series of frescoes in the Upper Church of Assisi a certain parallelism is to be found with works still to be seen at Rome. It is an obvious explanation of this parallelism to say that the artist who painted at Assisi was a pupil of the Roman school. The problem, however, requires more careful handling. It is true, indeed, that no document can be adduced to prove that Cavallini helped Giotto; equally true, however, that Giotto cannot be proved to have helped Cavallini. Yet certain facts of an important kind are known in regard to the status which Giotto, when he was at Rome, enjoyed. It is known that he designed and executed a mosaic for St Peter's, for which he received the enormous sum of