In the text of each of the chapters, of the Valladolid autograph there are omissions -- some merely verbal, often illustrating the author's aim in making the new redaction, others more fundamental. If the Valladolid manuscript represents the Way of perfection as St. Teresa wrote it in the period of her fullest powers, the greater freshness and individuality of the Escorial manuscript are engaging qualities, and there are many passages in it, omitted from the later version, which one would be sorry to sacrifice.
In what form, then, should the book be presented to English readers? It is not surprising if this question is difficult to answer, since varying procedures have been adopted for the presentation of it in Spain. Most of them amount briefly to a re-editing of the Valladolid manuscript. The first edition of the book, published at ƒvora in the year 1583, follows this manuscript, apparently using a copy (the so-called "Toledo" copy) made by Ana de San Pedro and corrected by St. Teresa; it contains a considerable number of errors, however, and omits one entire chapter -- the thirty-first, which deals with the Prayer of Quiet, a subject that was arousing some controversy at the time when the edition was being prepared. In 1585, a second edition, edited by Fray Jer—nimo Graci‡n, was published at Salamanca: the text of this follows that of the ƒvora edition very closely, as apparently does the text of a rare edition published at Valencia in 1586. When Fray Luis de Leon used the Valladolid manuscript as the foundation of his text (1588) he inserted for the first time paragraphs and phrases from that of El Escorial, as well as admitting variants from the copies corrected by the author: he is not careful however, to indicate how and where his edition differs from the manuscript.

Since 1588, most of the Spanish editions have followed Fray Luis de Le—n with greater or less exactness. The principal exception is the well-known "Biblioteca de Autores Espa–oles" edition, in which La Fuente followed a copy of the then almost forgotten Escorial manuscript, indicating in footnotes some of the variant readings in the codex of Valladolid. In the edition of 1883, the work of a Canon of Valladolid Cathedral, Francisco Herrero Bayona, the texts of the two manuscripts are reproduced in parallel columns. P. Silverio de Santa Teresa gives the place of honour to the Valladolid codex, on which he bases his text, showing only the principal variants of the Escorial manuscript but printing the Escorial text in full in an appendix as well as the text of the Toledo copy referred to above.

The first translations of this book into English, by Woodhead (1675: reprinted 1901) and Dalton (1852), were based, very naturally, on the text of Luis de Le—n, which in less critical ages than our own enjoyed great prestige and was considered quite authoritative. The edition published in 1911 by the Benedictines of Stanbrook, described on its title-page as "including all the variants" from both the Escorial and the Valladolid manuscript, uses Herrero Bayona and gives an eclectic text based on the two originals but with no indications as to which is which. The editors' original idea of using one text only, and showing variants in footnotes, was rejected in the belief that "such an arrangement would prove bewildering for the generality of readers" and that anyone who could claim the title of "student" would be able to read the original Spanish and would have access to the Herrero Bayona edition. Father Zimmerman, in his introduction, claimed that while the divergences between the manuscripts are sometimes "so great that the [Stanbrook] translation resembles a mosaic composed of a large number of small bits, skilfully combined", "the work has been done most conscientiously, and while nothing has been added to the text of the Saint, nothing has been omitted, except, of course, what would have been mere repetition".

This first edition of the Benedictines' translation furnished the general reader with an attractive version of what many consider St. Teresa's most attractive book, but soon after it was published a much more intelligent and scholarly interest began to be taken in the Spanish mystics and that not only by students with ready access to the Spanish original and ability to read it. So, when a new edition of the Stanbrook translation was called for, the editors decided to indicate the passages from the Escorial edition which had been embodied in the text by enclosing these in square brackets. In 1911, Father Zimmerman, suspecting that the procedure then adopted by the translators would not "meet with the approval of scholars", had justified it by their desire "to benefit the souls of the faithful rather than the intellect of the student"; but now, apparently, he thought it practicable to achieve both these aims at once. This resolution would certainly have had the support of St. Teresa, who in this very book describes intelligence as a useful staff to carry on the way of perfection. The careful comparison of two separate versions of such a work of genius may benefit the soul of an intelligent reader even more than the careful reading of a version compounded of both by someone else.

When I began to consider the preparation of the present translation it seemed to me that an attempt might be made to do a little more for the reader who combined intelligence with devoutness than had been done already. I had no hesitation about basing my version on the Valladolid MS., which is far the better of the two, whether we consider the aptness of its illustrations, the clarity of its expression, the logical development of its argument or its greater suitability for general reading. At the same time, no Teresan who has studied the Escorial text can fail to have an affection for it: its greater intimacy and spontaneity and its appeal to personal experience make it one of the most characteristic of all the Saint's writings -- indeed, excepting the Letters and a few chapters of the Foundations, it reveals her better than any. Passages from the Escorial MS. must therefore be given: thus far I followed the reasoning of the Stanbrook nuns.

Where this translation diverges from theirs is in the method of presentation. On the one hand I desired, as St. Teresa must have desired, that it should be essentially her mature revision of the book that should be read. For this reason I have been extremely conservative as to the interpolations admitted into the text itself: I have rejected, for example, the innumerable phrases which St. Teresa seems to have cut out in making her new redaction because they were trivial or repetitive, because they weaken rather than reinforce her argument, because they say what is better said elsewhere, because they summarize needlessly or because they are mere personal observations which interrupt the author's flow of thought, and sometimes, indeed, are irrelevant to it. I hope it is not impertinent to add that, in the close study which the adoption of this procedure has involved, I have acquired a respect and admiration for St. Teresa as a reviser, to whom, as far as I know, no one who has written upon her has done full justice. Her shrewdness, realism and complete lack of vanity make her an admirable editor of her own work, and, in debating whether or no to incorporate some phrase or passage in my text I have often asked myself: Would St. Teresa have included or omitted this if she had been making a fresh revision for a world-wide public over a period of centuries?"

At the same time, though admitting only a minimum of interpolations into my text, I have given the reader all the other important variants in footnotes. I cannot think, as Father Zimmerman apparently thought, that anyone can find the presence of a few notes at the foot of each page "bewildering". Those for whom they have no interest may ignore them; others, in studying them, may rest assured that the only variants not included (and this applies to the variants from the Toledo copy as well as from the Escorial MS.) are such as have no significance in a translation. I have been rather less meticulous here than in my edition of St. John of the Cross, where textual problems assumed greater importance. Thus, except where there has been some special reason for doing so, I have not recorded alterations in the order of clauses or words; the almost regular use by E. of the second person of the plural where V. has the first; the frequent and often apparently purposeless changes of tense; such substitutions, in the Valladolid redaction, as those of "Dios" or "Se–ior m’o" for "Se–ior"; or merely verbal paraphrases as (to take an example at random) "Todo esto que he dicho es para . . ." for "En todo esto que he dicho no trato . . ." Where I have given variants which may seem trivial (such as "hermanas" for "hijas", or the insertion of an explanatory word, like "digo") the reason is generally that there seems to me a possibility that some difference in tone is intended, or that the alternative phrase gives some slight turn to the thought which the phrase in the text does not.

The passages from the Escorial version which I have allowed into my text are printed in italics. Thus, without their being given undue prominence (and readers of the Authorized Version of the Bible will know how seldom they can recall what words are italicized even in the passages they know best) it is clear at a glance how much of the book was intended by its author to be read by a wider public than the nuns of St. Joseph's. The interpolations may be as brief as a single expressive word, or as long as a paragraph, or even a chapter: the original Chapter 17 of the Valladolid MS., for example, which contains the famous similitude of the Game of Chess, was torn out of the codex by its author (presumably with the idea that so secular an illustration was out of place) and has been restored from the Escorial MS. as part of Chapter 16 of this translation. No doubt the striking bullfight metaphor at the end of Chapter 39 was suppressed in the Valladolid codex for the same reason. With these omissions may be classed a number of minor ones -- of words or phrases which to the author may have seemed too intimate or colloquial but do not seem so to us. Other words and phrases have apparently been suppressed because St. Teresa thought them redundant, whereas a later reader finds that they make a definite contribution to the sense or give explicitness and detail to what would otherwise be vague, or even obscure. A few suppressions seem to have been due to pure oversight. For the omission of other passages it is difficult to find any reason, so good are they: the conclusion of Chapter 38 and the opening of Chapter 41 are cases in point.

The numbering of the chapters, it should be noted, follows neither of the two texts, but is that traditionally employed in the printed editions. The chapter headings are also drawn up on an eclectic basis, though here the Valladolid text is generally followed.

The system I have adopted not only assures the reader that he will be reading everything that St. Teresa wrote and nothing that she did not write, but that he can discern almost at a glance, what she meant to be read by her little group of nuns at St. Joseph's and also how she intended her work to appear in its more definitive form. Thus we can see her both as the companion and Mother and as the writer and Foundress. In both roles she is equally the Saint.

But it should be made clear that, while incorporating in my text all important passages from the Escorial draft omitted in that of Valladolid, I have thought it no part of my task to provide a complete translation of the Escorial draft alone, and that, therefore, in order to avoid the multiplication of footnotes, I have indicated only the principal places where some expression in the later draft is not to be found in the earlier. In other words, although, by omitting the italicized portions of my text, one will be able to have as exact a translation of the Valladolid version as it is possible to get, the translation of the Escorial draft will be only approximate. This is the sole concession I have made to the ordinary reader as opposed to the student, and it is hardly conceivable, I think, that any student to whom this could matter would be unable to read the original Spanish.

One final note is necessary on the important Toledo copy, the text of which P. Silverio also prints in full. This text I have collated with that of the Valladolid autograph, from which it derives. In it both St. Teresa herself and others have made corrections and additions -- more, in fact, than in any of the other copies extant. No attempt has been made here either to show what the Toledo copy omits or to include those of its corrections and additions -- by far the largest number of them -- which are merely verbal and unimportant, and many of which, indeed, could not be embodied in a translation at all. But the few additions which are really worth noting have been incorporated in the text (in square brackets so as to distinguish them from the Escorial additions) and all corrections which have seemed to me of any significance will be found in footnotes.